Orals
BrainWorld: A Structural-Prior-Conditioned Generative Model for Whole-Brain 4D fMRI Dynamics
Junfeng XIA ⋅ Wenhao Ye ⋅ Junxiang Zhang ⋅ Xuanye Pan ⋅ Mo WANG ⋅ Quanying Liu
Whole-brain 4D fMRI generation is valuable for modeling functional brain dynamics, yet existing fMRI foundation models mainly target representation learning and downstream prediction rather than conditional predictive generation. We introduce BrainWorld, a structural-prior-conditioned generative model for whole-brain 4D fMRI dynamics. BrainWorld uses sMRI as subject-level anatomical context to guide future fMRI generation, integrating structural information into the denoising process rather than treating it as a parallel modality. Evaluated on 22 datasets spanning diverse cohorts and brain states, BrainWorld generates stable 4D fMRI trajectories up to 400 frames, improves downstream performance through generated-example augmentation, and learns transferable multimodal representations that outperform baselines. Together, these results establish BrainWorld as a condition-aware generative framework for long-horizon brain dynamics modeling and multimodal representation learning.
Show more
Proactive Instance Navigation with Comparative Judgment for Ambiguous User Queries
Junhyuk Kwon ⋅ Seungjoon Lee ⋅ Hyejin Park ⋅ Kyle Min ⋅ Jungseul Ok
Natural-language instance navigation becomes challenging when the initial user request does not uniquely specify the target instance. A practical agent should reduce the user's burden by actively asking only the information needed to distinguish the intended target from similar distractors, rather than requiring a detailed description upfront. Existing approaches often fall short of this goal: they may stop at the first plausible candidate before sufficiently exploring alternatives, or, even after collecting multiple candidates, ask about the target's attributes derived from individual candidates rather than questions selected to distinguish candidates in the pool. As a result, despite the dialogue, the agent may still fail to distinguish the target from distractors, leading to premature decisions and lengthy user responses. We propose Proactive Instance Navigation with Comparative Judgment (ProCompNav), a two-stage framework that first constructs a candidate pool and then identifies the target through comparative judgment. At each round, ProCompNav extracts an attribute-value pair that splits the current pool, asks a binary yes/no question, and prunes all inconsistent candidates at once. This reframes disambiguation from open-ended target description to pool-level discriminative questioning, where each question is chosen to narrow the candidate set. On CoIN-Bench, ProCompNav improves Success Rate over interactive baselines with the same minimal input and non-interactive baselines with detailed descriptions, while substantially reducing Response Length. ProCompNav also achieves state-of-the-art Success Rate on TextNav, suggesting that comparative judgment is broadly useful for instance-level navigation among similar distractors.
Show more
CIG: Exploration via Conditional Information Gain
Tim Joseph ⋅ Marcus Fechner ⋅ Philipp Stegmaier ⋅ Karam Daaboul ⋅ Marius Zöllner
Intrinsic rewards for exploration in reinforcement learning condition on different contexts: lifelong rewards score each transition against accumulated experience but ignore within-rollout redundancy; episodic rewards penalize intra-trajectory repetition but discard lifetime progress. Hybrid methods combine both signals through heuristic weights or require Gaussian-process dynamics that do not scale beyond low-dimensional state spaces. Trajectory-level information gain decomposes into per-step terms that condition on the replay buffer and rollout prefix simultaneously, but remains intractable for deep models. We derive the Conditional Information Gain (CIG) reward as a tractable surrogate: a log-determinant objective over an ensemble disagreement kernel whose Cholesky factorization yields causal per-step rewards that retain both conditioning sets while scaling to high-dimensional state spaces. We instantiate CIG in a model-based setting, where rollouts are short and within-rollout corrections remain largely unexplored. Across twelve tasks spanning discrete (MiniGrid) and continuous control (OGBench), in both clean and stochastic-distractor settings, CIG outperforms or matches prior exploration methods while remaining robust to stochastic distractors.
Show more
LatentUM: Unleashing the Potential of Interleaved Cross-Modal Reasoning via a Latent-Space Unified Model
Jiachun Jin ⋅ Zetong Zhou ⋅ Xiao Yang ⋅ Hao Zhang ⋅ Pengfei Liu ⋅ Jun Zhu ⋅ Zhijie Deng
Unified models (UMs) hold promise for their ability to understand and generate content across heterogeneous modalities. Compared to merely generating visual content, the use of UMs for interleaved cross-modal reasoning is more promising and valuable, e.g., for solving understanding problems that require dense visual thinking, improving visual generation through self-reflection, or modeling visual dynamics of the physical world guided by stepwise action interventions. However, existing UMs necessitate pixel decoding as a bridge due to their disjoint visual representations for understanding and generation, which is both ineffective and computationally cumbersome. In this paper, we introduce LatentUM, a novel unified model that represents all modalities within a shared semantic latent space, eliminating pixel-space mediation for model-internal reasoning over generated visual states. By making generated visual tokens directly interpretable to the model itself, LatentUM enables flexible interleaved cross-modal reasoning and generation. Empirically, LatentUM achieves state-of-the-art performance on the Visual Spatial Planning benchmark, pushes the limits of visual generation through self-reflection and demonstrates world modeling capability by predicting future visual states in the shared semantic latent space.
Show more
LoRAcles: Self-Supervised Weight-Space Interpretability at Scale
Celeste De Schamphelaere ⋅ Jan Bauer ⋅ Neel Nanda ⋅ Euan Ong
Fine-tuned LLMs can learn complex and subtle behaviors. However, it can be difficult to ascertain what behaviors the model acquired during fine-tuning. We introduce LoRAcles: fine-tuned language models that take LoRA adapter weights as input, and answer natural language questions about them. We introduce a self-supervised pipeline for training LoRAcles: we first train LoRAs on small sets of pre-training documents, and then train LoRAcles on these LoRAs to answer questions about the documents. LoRAcles generalize far beyond their training data: they are state-of-the-art on AuditBench, a benchmark of models with sophisticated hidden behaviors; they can detect subtle changes in model organisms (fine-tunes deliberately constructed to encode a hidden behavior for auditing research), such as subliminal learning; and they are the first tool to achieve non-trivial performance at verbalizing semantic backdoor triggers. LoRAcles can also describe some learned behaviors from the weights of a full-parameter fine-tune. LoRAcles are a highly scalable method: we train LoRAcles for Qwen-3-14B and Llama-3.3-70B, and observe that performance scales smoothly with the size of our training dataset up to 100K LoRAs, though our method may need refinement in order to scale further. We note, though, that LoRAcles are prone to hallucinations and currently only elicit the most salient behavioral changes, limiting their utility beyond narrow fine-tunes.
Show more
IDEA: Unwrapping Visual Black-box Models by Interaction Decomposition
Chaojie Ji ⋅ Jian Xu ⋅ He Zhang ⋅ Hongyan Wu ⋅ Yankai Cao ⋅ Ruxin Wang
Concept-based explanation methods have emerged as a prominent framework for interpreting deep neural networks using human-understandable concepts. However, existing approaches are limited to assigning a scalar importance score to a concept, failing to capture the interaction between the concept and its visual context. We argue that overlooking this interaction leaves a predictor's reasoning only partially understood. To address this gap, we introduce IDEA, a novel global post-hoc explanation framework that explicitly models the interaction between a concept and its visual context. Through this modeling, IDEA decomposes concept-level predictive information into three atomic components: uniqueness (independent concept contribution), redundancy (information shared with the context), and synergy (information arising strictly from concept-context interaction). Consequently, IDEA extends interpretability beyond merely identifying influential concepts to diagnosing how a predictor reasons -- for instance, flagging potential context dependencies that scalar scores cannot reveal. IDEA requires no task-specific concept annotations and no access to classifier internals. Evaluations on controlled and real-world settings demonstrate that IDEA can surface reasoning patterns invisible to traditional scalar concept scores.
Show more
Thinking with Images as Continuous Policy: Numerical Visual Chain-of-Thought
Kesen Zhao ⋅ Beier Zhu ⋅ Junbao Zhou ⋅ Xingyu Zhu ⋅ Zhongqi Yue ⋅ Hanwang Zhang
Recent multimodal large language models (MLLMs) increasingly rely on visual chain-of-thought to perform region-grounded reasoning over images. However, existing approaches ground regions via either textified coordinates—causing modality mismatch and semantic fragmentation—or fixed-granularity patches that both limit precise region selection and often require non-trivial architectural changes. In this paper, we propose Numerical Visual Chain-of-Thought (NV-CoT), a framework that enables MLLMs to reason over images using continuous numerical coordinates. NV-CoT expands the MLLM action space from discrete vocabulary tokens to a continuous Euclidean space, allowing models to directly generate bounding-box coordinates as actions with only minimal architectural modification. The framework supports both supervised fine-tuning and reinforcement learning. In particular, we replace categorical token policies with a Gaussian (or Laplace) policy over coordinates and introduce stochasticity via reparameterized sampling, making NV-CoT fully compatible with GRPO-style policy optimization. Extensive experiments on three benchmarks against eight representative visual reasoning baselines demonstrate that NV-CoT significantly improves localization precision and final answer accuracy, while also accelerating training convergence, validating the effectiveness of continuous-action visual reasoning in MLLMs. The code is provided in the Supplementary Materials.
Show more
Can In-Context Learning Support Intrinsic Curiosity?
Eric Elmoznino ⋅ Sangnie Bhardwaj ⋅ Johannes von Oswald ⋅ Rajai Nasser ⋅ Blaise Aguera y Arcas ⋅ João Sacramento ⋅ Rif A. Saurous ⋅ Guillaume Lajoie
Effective machine learning depends not only on how we model data, but also on what data we choose to collect. While large sequence models have revolutionized data modeling, the problem of automated data selection, or "intrinsic curiosity", remains a significant challenge. Classic approaches incentivize exploration by rewarding an agent based on its "learning progress", which measures how much a newly acquired observation improves a world model's predictive ability. However, evaluating these rewards traditionally requires expensive inner loops of gradient descent updates within each trajectory, rendering them computationally impractical at scale. In this work, we investigate whether the emergent in-context learning (ICL) capabilities of sequence models can eliminate this bottleneck by serving as immediate, update-free world models. Specifically, we evaluate whether an exploration policy can be trained to maximize learning progress, using solely the prediction errors and counterfactual context manipulations of an in-context learner. We first prove that in general Markov decision processes, this is in fact impossible in an unbiased way: the resulting intrinsic rewards either suffer from nuisance terms that bias their estimation of true learning progress, or they cannot be implemented using an in-context learner's prediction errors. Conversely, we prove a positive result for a broad subclass of non-temporal settings, encompassing active learning and Bayesian Experimental Design: here, ICL-derived rewards successfully bound and asymptotically converge to the true learning progress. We corroborate our theory with controlled experiments across continuous and symbolic environments, demonstrating that our ICL-driven framework successfully trains curious data-collection policies that explore optimally.
Show more
PhaseLoRA: Control-Regime-Conditioned Low-Rank Adaptation for Continuous-Action Vision-Language-Action Policies
Yufei Guo ⋅ Yinan Wu ⋅ Haoran Duan ⋅ guiguang ding ⋅ Jungong Han
Parameter-efficient fine-tuning (PEFT) is a natural way to adapt pretrained vision-language-action (VLA) policies, but most adapter designs apply temporally static updates throughout a control rollout, overlooking the phase-dependent nature of continuous-action manipulation. Such policies traverse distinct regimes, including approach, contact transition, grasping, transport, and placement, each requiring different adaptation behaviors. We propose \textbf{PhaseLoRA}, a lightweight LoRA parameterization that conditions adaptation at each control step using two weakly supervised descriptors: fine-control tendency and event/boundary intensity. PhaseLoRA modulates the LoRA left factor in the action expert, allowing the effective low-rank update direction to vary over time while keeping the backbone largely frozen. On LIBERO, PhaseLoRA improves average success rate by 12.2 points over a matched-parameter high-rank LoRA baseline and outperforms stronger LoRA variants. Ablations and update-direction analyses show that the gain is not explained by parameter count, arbitrary temporal modulation, or scalar gating, but by control-regime-aligned changes in low-rank update directions. These results establish within-trajectory conditioning as an effective lightweight PEFT axis for continuous-action VLA policies.
Show more
A multi-scale information geometry reveals the structure of mutual information in neural populations
Simone Azeglio ⋅ Steeve Laquitaine ⋅ Ulisse Ferrari ⋅ Matthew Chalk
Understanding how neural population responses represent sensory information is a central problem in systems neuroscience. One approach is to define a representational geometry on stimulus space in which distances reflect how reliably stimuli can be distinguished from neural activity. However, different constructions of these distances can lead to qualitatively different conclusions about the neural code. Here, we show that a unique Riemannian representational geometry emerges from first principles governing how distances contract as stimulus resolution is lost through coarse-graining. This results in a multi-scale extension of the Fisher information metric, capturing encoding structure from fine stimulus details to coarse global distinctions. The resulting geometry is exactly related to the mutual information encoded by the population: well encoded stimulus directions -- those contributing more to mutual information -- are expanded, whereas poorly encoded directions are contracted. The metric tensor can be estimated using diffusion models, making the framework practical for large neural populations and high-dimensional stimuli. Applied to visual cortical responses to natural images, the eigenvectors of the metric tensor identify stimulus variations that contribute most to information transmission, yielding interpretable features that are robust to modelling choices. Together, these results provide a principled, information-theoretic framework for characterising neural population codes.
Show more
Neural population tuning statistics as priors for multitask generalization
Zach Cohen ⋅ Jan Drugowitsch
Neural populations exhibit heterogeneous tuning: Neurons differ in their selectivity, gain, and in how their preferred inputs are distributed across stimulus or latent-variable space. While this heterogeneity has been studied extensively in the context of encoding precision, its implications for learning and generalization are less well-understood. Here, we develop a probabilistic theory that links a neural population's tuning statistics to its inductive bias: in large populations, the tuning statistics induce a kernel and, equivalently, a prior of a generative model over functions. This prior in turn predicts the population's ability to generalize across a distribution of related tasks. Our theory determines the population tuning statistics that minimize the generalization error over a distribution of functions. We further present a normative account of tuning adaptation inspired by our probabilistic approach, showing that tuning changes that maximize the marginal likelihood of observed data systematically reduce generalization error relative to non-adapting populations. Applied to the adaptation of hippocampal spatial representations in a reward learning task, we found that our theory could explain the over-representation of rewarding locations through experience in an environment with a localized reward. We further demonstrate how different components of the population's generative model differentially impact the speed and magnitude of tuning adaptation. Overall, our work proposes a novel connection between tuning variability and a population's ability to learn and to generalize across tasks and develops a Bayesian theory of tuning adaptation in neural populations during learning.
Show more
Emergence of Physical Intelligence via Controllable Information Production
Tristan Shah ⋅ Stas Tiomkin
Intrinsic Motivation (IM) aims to train agents without external rewards, enabling useful behavior to emerge from the agent's interaction with its environment alone. However, the dominant IM approaches rely on information-theoretic quantities with designer-chosen variables, introducing bias and lacking a principled connection to dynamics or optimal control (OC). We introduce Controllable Information Production (CIP), a new foundation for IM explicitly grounded in dynamical systems and OC. CIP measures the rate at which an agent produces information, capturing controllable complexity without external knowledge or bias. CIP unifies IM and OC into a single framework, formalizing physical intelligence as the control of information production. It further reveals connections between the structure of the value function and Kolmogorov–Sinai entropy. CIP consistently outperforms prior IM methods on standard benchmarks in robot learning and solves tasks they fail on, including humanoid self-righting. These results support a general organizing principle: physical intelligence emerges from driving systems toward the edge of controllable chaos.
Show more
Decomposing Earth Embeddings with Sparse Autoencoders
Vitus Benson ⋅ Fanny Yang ⋅ Markus Reichstein
Earth observation foundation models produce global meter-resolution embeddings that underpin work across agriculture, ecology, hydrology, and beyond. Yet the embeddings themselves are opaque: each dimension entangles unrelated physical concepts, rendering their interpretability and robustness challenging for practitioners and scientists alike. Here, we propose sparse autoencoders (SAEs) as a principled way to decompose Earth embeddings into overcomplete dictionaries of sparse, approximately monosemantic features. We introduce an evaluation protocol suited to a domain where LLM-based auto-interpretability fails and spatial autocorrelation matters, scoring any SAE on reconstruction, sparsity, monosemantic recovery, and spatial coherence against a curated label registry of categorical and continuous reference products. On this protocol, three state-of-the-art SAE recipes (ReLU+$\ell_1$, BatchTopK, and Matryoshka BatchTopK) all recover monosemantic and spatially contiguous concepts from AlphaEarth and Tessera embeddings much better than raw embeddings, with Matryoshka the strongest of the three. Two applications follow: (i) unsupervised discovery of land-cover subconcepts (e.g., splitting water into river, lake, ocean), and (ii) improved area-of-applicability estimates for OOD detection. Together, these results are a step toward Earth observation models whose features can be inspected rather than only used, opening pathways for scientific discovery and improved trustworthiness in geospatial applications. Code and models will be released upon publication.
Show more
They Can See It but not Say It: Iterative Self Knowledge Re-expression in Visual Reasoning Relative Pose Identification
MENGYU WANG ⋅ Ken Deng ⋅ Yftah Ziser ⋅ Shay Cohen ⋅ Tiejun Ma
The task of visual reasoning relative pose identification (VRRPI) has exposed critical weaknesses in the multi-view 3D spatial reasoning capabilities of modern vision-language models (VLMs). We present a systematic analysis of VLMs behavior in VRRPI, uncovering stable output biases where VLMs disproportionately predict specific motion patterns regardless of visual evidence. To mitigate these inherent biases, we propose self knowledge re-expression with iterative debiasing (SKR-ID), an annotation-free pipeline that adapts VLMs for VRRPI using only randomly extracted unannotated image pairs. SKR-ID transitions VLMs' output mechanism from generic next-token prediction to task-optimized classification heads. The pipeline proceeds in two stages: (1) a cold-start initialization using debiasing object-centric prompts to anchor motion inference in 2D visual evidence; and (2) a recursive logit-ranking mechanism that enforces statistical parity by using the median logit value as a dynamic threshold. Experimental evaluations across frontier VLMs on standard benchmarks demonstrate that SKR-ID significantly mitigates systematic biases and unearths latent 3D reasoning ability. Our method achieves substantial gains, improving open-sourced VLMs' Macro-F1 scores by at least 8.3% and at most 23.0%, proving that VLMs possess significant untapped multi-view 3D reasoning potential.
Show more
Resilience Matters for Embodied Agents System: New Metrics, Systematic Evaluation, and Optimization
Yapeng Liu ⋅ Yuanzhao Zhai ⋅ Xudong Gong ⋅ Feng Dawei ⋅ Bo Ding ⋅ Lin Wang ⋅ Huaimin Wang
As Embodied Agents System (EAS) move into physical domains such as autonomous navigation and household assistance, reliable execution becomes critical for physical task completion and collaboration trust. Recent EAS reliability evaluation works typically focus on outcome-centric metrics as success rate or safety-only scores to assess an agent's performance, which collapse diverse execution trajectories into coarse outcomes. Therefore they ignore a critical property of EAS -- which we define as the **Resilience** -- that reflects how EASs recover, stabilize, and extend under perturbations and across iterative updates. The lack of resilience is particularly critical in open-world environments due to continuous unexpected disruptions, thus directly affecting the quality of EAS deployment. To address this problem, we gain insight from the resilience-engineering concepts to EAS groundings and propose a novel resilience evaluation framework that can be flexibly applied to any EAS. Specifically, we define the first comprehensive resilience metrics suite for EASs system that exposes *Rebound*, *Stability*, and *Graceful Extensibility* across embodied tasks execution, providing a practical grounding for EAS resilience analysis. We further implement the resilience evaluation layer that transforms execution evidence into assessments for comparison, diagnosis and optimization. Across 400 household tasks with 10 representative EAS baselines, our evaluation reveals process-level distinction hidden by outcome metrics, including recovery cost differences among successful episodes ($\Delta C_{\mathrm{rec}}=25.2$), increased instability under semantic perturbations, and task-family-specific degradation under stress. Metrics-guided optimizations reduce recovery cost by 42.94%, increase stability by 19.87%, and improve graceful extensibility completion by 10.08%, showing the diagnostic effect of resilience evaluation. Beyond these targeted gains, our results reveal trade-offs among resilience characteristics, suggesting that a resilient EAS construction should be evaluated and configured according to deployment-specific requirements. Our code and data are available at .
Show more
GeoSym127K: Scalable Symbolically-verifiable Synthesis for Multimodal Geometric Reasoning
Jinhao Jing ⋅ Zheng Ma ⋅ Jinwei Liang ⋅ Qiannian Zhao ⋅ Shuang Chen ⋅ Jing Yang ⋅ Prayag Tiwari ⋅ Jingjing Bai ⋅ Benyou Wang ⋅ Por L Yee ⋅ Zhan Su ⋅ Lewei Lu
Large Multimodal Models (LMMs) often struggle with geometric reasoning due to visual hallucinations and a lack of mathematically precise Chain-of-Thought (CoT) data. To address this, we propose the GeoSym Engine, an automated and scalable neuro-symbolic framework. By leveraging a type-conditional grammar and an analytic SymGT Solver, it derives exact symbolic ground truths and seamlessly integrates with a robust rendering pipeline to produce high-precision geometric diagrams. Using this engine, we construct GeoSym127K, a difficulty-stratified dataset featuring 51K high-resolution images, 127K questions with symbolic ground truths, and 55K answer-verified CoT QA pairs. We also introduce GeoSym-Bench, an expert-curated suite of 511 complex samples for rigorous evaluation. Through extensive supervised fine-tuning (SFT), we demonstrate that GeoSym drives concentrated improvements specifically on diagram-dependent and multi-step geometry tasks. Our Qwen3-VL-8B model gains an absolute +22.21% on the MathVerse Vision-Only subset and reaches 61.52% (+6.19% improvement) on WeMath, mitigating long-horizon logic fragmentation and outperforming advanced closed-source models like Doubao-1.8. Furthermore, applying Reinforcement Learning with Verifiable Rewards (RLVR) via GRPO reveals that initializing from structural SFT checkpoints substantially elevates the performance ceiling over zero-shot RL. Driven by deterministic exact-match signals, this showcases the robust scaling potential of our verifiable reasoning synthesis. Datasets and code are available at https://huggingface.co/datasets/Tomie0506/GeoSym127K and https://github.com/Tomie56/GeoSym127K
Show more
HeMeR: Heterogeneous Memory Reconciliation for Embodied Agents via Structured KV Reuse
Saehun Chun ⋅ Wonje Choi ⋅ Jinwoo Jang ⋅ TaeYoon Kwack ⋅ Honguk Woo
Embodied agents make decisions by combining information from heterogeneous memory modules, including declarative memory for explicit knowledge, procedural memory for learned routines, and working memory for current observations. During interaction, the memory contents required for decision making change with the agent's state, goal, and task progress, while many contents remain shared across consecutive decisions. However, existing memory modules are typically accessed through separate interfaces, causing each updated memory context to be processed as a new input to the foundation model. This prevents selective key-value (KV) cache reuse for unchanged memory contents and forces full KV cache recomputation even when much of the memory context remains valid. We introduce HeMeR, an inference framework that performs KV reconciliation between dynamically updated memory contexts and reusable KV states. HeMeR tracks declarative, procedural, and working memory contents across action decisions, preserves KV states for unchanged contents, and reconciles only the states affected by memory-content updates. This enables efficient and stable inference as memory contents evolve during interaction. We evaluate HeMeR on ALFWorld, Habitat, and real-world robotic deployments requiring coordinated use of heterogeneous memory contents. On Real-world, HeMeR achieves $83.20\%$ task success and reduces per-step inference time from $2.94$ to $1.87$ seconds compared with MemRL by preserving compatible KV states as task progress updates the memory context.
Show more
Coordination Connectivity: Shared Initialization Shapes the Joint-Policy Landscape in MARL
Yixiang Fan ⋅ Zhiqiang Pu ⋅ Hao Ma ⋅ Dongmin Li ⋅ Weihao Zhang ⋅ Yinxing Dai ⋅ Shijie Zheng ⋅ Jingjing Huang
Mode connectivity studies reveal that high-performing neural networks are often connected by low-loss paths in parameter space. We investigate the analogue of this phenomenon in cooperative multi-agent reinforcement learning (MARL), where a solution is not a single model but a team of decentralized policies. We introduce coordination connectivity, a diagnostic evaluating whether linear interpolation between the actor parameters of two trained policy teams preserves high team return. The resulting coordination barrier measures whether two successful teams are separated by a performance cliff along the joint-policy path. We evaluate this diagnostic on the StarCraft Multi-Agent Challenge (SMAC) across five controlled training histories, isolating the effects of shared initialization, continued training, parameter decoupling, and HAPPO-style separate-actor specialization. Across three SMAC maps and four seeds, policy teams derived from a shared MAPPO checkpoint exhibit near-zero interpolation barriers, even after non-shared specialization. In contrast, interpolation paths involving independently trained HAPPO teams exhibit substantially larger barriers. Mixed-team assembly, single-agent hot-swap, clone-all, role-swap, and weight-space probes further indicate that shared-initialized agents specialize into distinct roles while remaining within a low-barrier coordination component. These results suggest that shared initialization implicitly structures the joint-policy space of cooperative MARL, providing a mode-connectivity-inspired lens for analyzing coordination, specialization, and same-source cross-version team compatibility.
Show more
Learning Sparse Semantic-Cortical Atoms for Multisubject Naturalistic fMRI Encoding
Alicia Zeng
Naturalistic language encoding models based on large language model features can predict cortical responses well, but their learned representations remain difficult to interpret. Three challenges are central: semantic selectivity is fundamentally non-identifiable in high-dimensional correlated feature spaces, cortical organization remains implicit in voxelwise weight maps, and cross-subject structure is usually addressed only after model fitting. We introduce a multisubject coupled sparse autoencoder that jointly reconstructs language-model features and predicts subject-specific brain responses, learning a sparse basis of semantic-cortical atoms that makes semantic structure, cortical organization, and cross-subject correspondence explicit within a single model. In naturalistic story-listening fMRI, the model retains 88\% of matched dense-feature ridge brain prediction performance using 50 atoms and $k{=}5$ active atoms per TR. Relative to matched two-stage baselines, joint training yields the strongest overall tradeoff between prediction and interpretability. The learned atoms are semantically coherent and interpretable, recover more reproducible cross-subject brain maps than standard post hoc encoding-model analyses, align with large-scale cortical structure from Yeo~7 and Neurosynth, and transfer better than ridge to held-out subjects in the low-data regime. This framework provides a more interpretable basis for cognitive neuroscience analyses of semantic-to-brain encoding by linking semantic features to corresponding cortical response patterns across subjects.
Show more
Learning Reveals Invisible Structure in Low-Rank RNNs
Yoav Ger ⋅ Omri Barak
Learning in neural systems arises from synaptic changes that reshape the representations underlying behavior. While low-rank recurrent neural networks (RNNs) have emerged as a powerful framework for linking connectivity to function, a theoretical understanding of their learning process remains elusive. Here, we extend the low-rank framework from activity to learning by deriving gradient-descent dynamics directly in a reduced overlap space. We formulate a closed-form, low-dimensional system of ODEs that governs learning in this space, exact for linear RNNs and asymptotically exact for nonlinear RNNs in the large-$N$ Gaussian limit. Central to our analysis is a distinction between two classes of overlaps: _loss-visible_ overlaps, which fully determine network activity, output, and loss, and _loss-invisible_ overlaps, which do not affect function but are required to describe learning. We illustrate the consequences of this decomposition through two phenomena. First, we show that learning can serve as a perturbation that exposes differences in connectivity between functionally equivalent networks. Second, we show that loss-invisible overlaps can act as memory variables that encode training history, and characterize the conditions under which this occurs. Finally, we present several testable predictions for biological learning experiments derived from our theory.
Show more
HABIT: Human-Aware Behavior and Interaction Training Dataset for Robot Manipulation
Jaehwi Song ⋅ Suchae Jeong ⋅ Byeongguk Jeon ⋅ Sungdong Kim ⋅ Hyungmok Son ⋅ Kimin Lee
Large-scale demonstration datasets have been central to recent progress in general-purpose robot policies. However, existing datasets are collected in human-absent settings, and policies trained on such data may perform tasks competently in isolation but fail to exhibit human-aware behaviors. To address this gap, we introduce HABIT, a large-scale robot demonstration dataset for human-present environments. We organize tasks into three roles capturing distinct modes of human-robot interaction: Collaborator, where human and robot jointly accomplish a task; Coworker, where they pursue separate tasks in a shared space; and Supervisor, where the human directs the robot. The dataset comprises over 7K episodes and over 100 hours across 48 tasks. Our experiments show that training on human-present data elicits human-aware behaviors that robot-only data fails to produce: spatiotemporal synchronization in Collaborator tasks, yielding in Coworker tasks, and gesture grounding in Supervisor tasks. Moreover, training on HABIT enables rapid adaptation to new human-robot interaction tasks. By introducing human presence as a new axis of dataset diversity, HABIT extends robot policies to environments shared with humans.
Show more
Clarification as Supervision: Reinforcement Learning for Vision-Language Interfaces
John Gkountouras ⋅ Ivan Titov
Recent text-only models demonstrate remarkable reasoning capabilities. Extending these to visual domains requires vision-language models to translate images into text descriptions. However, current models, trained to produce captions for human readers, often omit the precise details that reasoning systems require. This creates an interface mismatch: reasoners often fail not due to reasoning limitations but because they lack access to critical visual information. We propose Adaptive-Clarification Reinforcement Learning (AC-RL), which, through interaction, teaches vision models what information reasoners need. Our key insight is that clarification requests during training reveal information gaps; by penalizing success that requires clarification, we create pressure for comprehensive initial captions that enable the reasoner to solve the problem in a single pass. AC-RL improves average accuracy by $4.4$ points over pretrained baselines across seven visual reasoning benchmarks, and analysis shows it would cut clarification requests by up to $44$\% if those were allowed. By treating clarification as a form of implicit supervision, AC-RL demonstrates that vision-language interfaces can be effectively learned through interaction alone, without requiring explicit annotations.
Show more
Geometry-Aware Similarity Metrics for Neural Representations on Riemannian and Statistical Manifolds
N Alex Cayco Gajic ⋅ Arthur Pellegrino
Similarity measures such as canonical correlation analysis (CCA), representational similarity analysis (RSA) and center kernel alignment (CKA), are widely used to compare the representational geometries used by different neural networks to solve the same task. Yet, because existing methods compare the *extrinsic geometry* of neural representations in state space, rather than their *intrinsic geometry*, they may fail to capture subtle yet crucial distinctions between fundamentally different computations performed by biological and artificial neural networks. Here, we introduce metric similarity analysis (MSA), a novel method which leverages tools from Riemannian geometry to compare the intrinsic geometry of neural manifolds. Within our mathematical framework, we derive several properties of MSA, including coordinate, scale and rotation invariances. We show that MSA can be used to i) disentangle neural computations of deep networks in rich vs. lazy learning regimes, ii) compare the computation-through-dynamics of recurrent neural networks and state-space models during neuroscience working memory tasks, and iii) investigate the statistical manifolds of diffusion models. Overall, we introduce a mathematically grounded and broadly applicable framework to understand the mechanisms behind neural computations by contrasting their Riemannian geometries, with broad applicability to both neuroscience and machine learning.
Show more
SMART: Scalable Multi-Agent Role-conditioned Teaming via LLM-free Tree Search
Runzhe Zhang ⋅ Ning Li ⋅ Xingshan Zeng ⋅ Letian Chen ⋅ Yasheng Wang ⋅ Weinan Zhang ⋅ Yong Yu ⋅ Weiwen Liu
Large language model (LLM)-powered multi-agent systems have been shown to effectively solve real-world end-to-end problems through collaboration among diverse agents. However, as LLM capabilities advance and user demands continue to grow, the range of problems that agents can address expands dramatically, while the population of available agents and tools continues to increase, giving rise to open and evolving ecosystems of autonomous agents. In this paradigm, agents are likely to become increasingly specialized, each focusing on particular capabilities or problem types. In such a setting, a central challenge is how to efficiently assemble an appropriate team of agents to take on different responsibilities from a massive agent pool for a given task. In this work, we study this problem in a controlled setting with 300 task-oriented agents. SMART (Scalable Multi-Agent Role-conditioned Teaming) first plans a task-conditioned collaboration topology and candidate role set, then constructs a compact role-conditioned candidate pool, and finally uses a LLM-free Monte Carlo Tree Search (MCTS) to assign agents to roles online via metadata-based rollouts. This design enables high flexibility without incurring prohibitive inference costs. Across multiple experiments, our approach achieves superior performance on GSM8K (93.64\%), MATH (61.73\%), HumanEval (93.89\%), and MBPP (87.68\%) using gpt-4o-mini under a limited budget.
Show more
MindAlign: Bridging EEG, Vision, and Language for Zero-Shot Visual Decoding
Zexuan Chen ⋅ Sichao Liu ⋅ Runhao Lu ⋅ Huichao Qi ⋅ Alexandra Woolgar ⋅ Xi V Wang ⋅ Lihui Wang
Visual decoding from brain signals is a key challenge at the intersection of computer vision and neuroscience, requiring methods that bridge neural representations and computational models of vision. A field-wide goal is to achieve accurate, generalizable decoding from non-invasive, temporally resolved signals, including electroencephalography (EEG). A major obstacle towards this goal is the low signal-to-noise ratio of EEG and the substantial inter-subject variability, which render direct end-to-end EEG–image supervision weak and unstable. To address this, we introduce a tri-modal contrastive framework for EEG-based visual decoding that aligns EEG, visual, and textual representations within a unified latent space. Our approach follows a two-stage design. First, we pre-train an EEG encoder via masked reconstruction on unlabeled trials, learning spatio-temporal regularities that transfer robustly downstream. Second, we jointly align EEG, image, and LLM-generated textual descriptions through contrastive learning, where text supervision acts as a semantic regularizer that injects linguistic structure into the shared space without overwhelming the primary EEG–image signal. The encoder integrates subject-specific adaptation, graph-attention over channels, and temporal-spatial convolutional embeddings. On the Things-EEG2 200-way zero-shot benchmark, our framework achieves 54.1\% Top-1 and 83.4\% Top-5 accuracy, substantially exceeding the strongest prior baseline (32.4\% / 64.0\%), with paired Wilcoxon tests confirming significance (p < 0.01) over all in-subject baselines. We validate generalization on Things-MEG. Analysis reveals that compact embedding geometries (CN-CLIP) outperform much larger backbones, and that decoding aligns with established neurophysiology of visual processing. This work is a critical step towards robust, semantically-grounded visual decoding from non-invasive temporal neural signals.
Show more
Evaluating Neural Data Tokenizers: A Framework for Assessing Learned Representations of Spiking Activity
Federico D'Agostino ⋅ Alex Gilbert ⋅ Susanne Keller ⋅ Jaivardhan Kapoor ⋅ Nicolas Reategui ⋅ Vedang Lad ⋅ Matthias Kümmerer ⋅ Jai Bhagat ⋅ Emin Orhan ⋅ Hasan A Bedel ⋅ Nikos Karantzas ⋅ Fabian Sinz ⋅ Surya Ganguli ⋅ Jakob H Macke ⋅ Sophia Sanborn ⋅ Alexander Ecker ⋅ Katrin Franke ⋅ Matthias Bethge ⋅ Andreas Tolias ⋅ Konstantin Willeke
Foundation models of brain activity aim to learn representations from large-scale neural recordings that generalize across sessions, subjects, species, and downstream tasks. Transformer-based models have emerged as a promising architecture for this goal and begin all with the same first step --- tokenization: transforming sparse binary events into vector representations that downstream networks can consume. In other areas of machine learning, tokenization is its own design problem, distinct from the downstream foundation model. Several mostly implicit tokenization approaches dominate the literature, but their impact on downstream performance has not been systematically evaluated. We propose four explicit desiderata for a neural tokenizer---reconstruction fidelity, downstream utility, cross-instance generalization, and compression---and a framework of metrics spanning seven dimensions, including firing statistics, encoding and decoding, and latent structure and robustness. We assemble a benchmark of ten datasets spanning species, brain areas, recording technologies, and behavioral paradigms, and evaluate a representative set of discrete and continuous tokenizers, both single-neuron and population variants. Across the strategies evaluated, we show how tokenizers trained only on reconstruction can support downstream tasks. Yet reconstruction alone is misleading: reconstruction quality and downstream utility can be sharply dissociated. Our multi-axis framework is a response to this failure mode. These findings highlight tokenization as a critical and underexplored design axis in brain foundation models, and suggest that further progress will require continued systematic study of tokenization strategies.
Show more
Incentivizing Medical Vision Capabilities from Large-Scale Multimodal Pre-training
Junying Chen ⋅ Zhenyang Cai ⋅ Yunjin Yang ⋅ Kunyuan Cai ⋅ Rongsheng Wang ⋅ Xidong Wang ⋅ Geyang Yu ⋅ Xiangyi Feng ⋅ Benyou Wang
Building capable medical vision-language models (VLMs) is challenging due to the scarcity of domain data and the difficulty of eliciting pre-trained medical visual knowledge. We present HuatuoGPT-5, a medical VLM built on a simple yet effective recipe: large-scale multimodal pre-training followed by reinforcement learning (RL). We first construct PubMedVision-Plus, the largest medical image-text dataset with 60M pairs from public PubMed literature, along with a clean 12M high-quality subset. After pre-training on this data, we find that traditional supervised fine-tuning (SFT) struggles to elicit the pre-trained knowledge. In contrast, by applying RL with verifiable and rubric-based instances synthesized directly from the pre-training data, we successfully incentivize the model to use this pre-trained medical visual knowledge. On PubMed-derived data, HuatuoGPT-5 outperforms similarly sized VLMs on multiple medical benchmarks and expert evaluations, significantly narrowing the gap between medical VLMs and leading proprietary VLMs. We hope these data, models, and findings can help accelerate open research on capable medical VLMs.
Show more
FLIP: Fast and Accurate Global Lipschitz Estimation for Large Feedforward Networks
Hongbo Chen ⋅ Zihao Ren
The Lipschitz constant of neural networks is widely used in robustness, control, generalization bounds, and training stability. As modern neural networks continue to scale, existing methods for Lipschitz estimation are either very time-consuming or overly loose. In this work, we focus on both fast and accurate Lipschitz estimation for large feedforward neural networks. We first reformulate the well-known semidefinite program (SDP) framework into an explicit recursive form, and introduce a key theoretical result, bound backpropagation: the SDP objective has upper and lower bounds that can be propagated backward along the recursive computation graph and depend only on partial variables. Leveraging this property, we decompose the original SDP into a sequence of small subproblems, each minimizing a combination of upper and lower bounds over partial variables. We prove that each subproblem is one-dimensional and convex, enabling fast solution via the secant method. Finally, we present the full FLIP algorithm and its complexity, yielding fast and tight Lipschitz estimation for large feedforward networks. Experiments on both randomly generated and trained networks show that FLIP improves speed and tightness by several orders of magnitude over prior methods, giving the tightest Lipschitz estimates for 100M-parameter networks in 3 seconds.
Show more
SAM 3D Animal: Promptable Animal 3D Reconstruction from Images in the Wild
Xuyi Hu ⋅ Jin Lyu ⋅ Jiuming Liu ⋅ Yebin Liu ⋅ Silvia Zuffi ⋅ Liang An ⋅ Stefan Goetz
3D animal reconstruction in the wild remains challenging due to large species variation, frequent occlusions, and the prevalence of multi-animal scenes, while existing methods predominantly focus on single-animal settings. We present SAM 3D Animal, the first promptable framework for multi-animal 3D reconstruction from a single image. Built on the SMAL+ parametric animal model, our method jointly reconstructs multiple instances and supports flexible prompts in the form of keypoints and masks which enable more reliable disambiguation in crowded and occluded scenes. To train such a model, we further introduce Herd3D, a multi-animal 3D dataset containing over 5K images, designed to increase diversity in species, interactions, and occlusion patterns. Experiments on the Animal3D, APTv2, and Animal Kingdom datasets show that our framework achieves state-of-the-art results over both existing model-based and model-free methods, demonstrating a scalable and effective solution for prompt-driven animal 3D reconstruction in the wild.
Show more
Bridging Compute- and Data-Optimal Pretraining
Tian Qin ⋅ Kimia Hamidieh ⋅ David Alvarez-Melis
Classical compute-optimal scaling laws assume an unbounded supply of fresh pretraining data, yet pretraining is increasingly entering the regime where compute is growing faster than high-quality data. We propose *Compute-Data (CD) scaling laws*, a unified framework that bridges *compute-optimal* scaling, in which data scales freely with compute, and *data-optimal* scaling, in which the corpus is fixed and compute can grow unbounded. CD scaling extends classic scaling by introducing a *token effectiveness function* that quantifies how much a *derived* token, produced for instance by multi-epoch repetition or paraphrasing, is worth relative to a fresh one, ranging from a perfect substitute to no value at all. Fitting $\eta$ for two data expansion strategies (multi-epoch repetition, paraphrasing) across model sizes ranging from 14M to 600M parameters on the Dolma-3 corpus, we find that it is far from constant: it depends jointly on model size, the tokens-per-parameter ratio, and the amount of derived data, and saturates as the corpus is expanded. The functional form of $\eta$ implies that substituting compute for data *diminishes* with both model size and data availability, and partitions training into three operational regimes: compute-bound, data-bound, and model-bound, showing that classic compute-optimal allocation is suboptimal across most of the practically relevant regime.
Show more
Do We Need Asynchronous SGD? On the Near-Optimality of Synchronous Methods
Grigory Begunov ⋅ Alexander Tyurin
Modern distributed optimization methods mostly rely on traditional synchronous approaches, despite substantial recent progress in asynchronous optimization. We revisit Synchronous SGD and its robust variant, called $m$-Synchronous SGD, which can be interpreted as a method with a backup-worker strategy, and theoretically show that they are nearly optimal in many heterogeneous computation and communication scenarios, which is somewhat unexpected. We analyze the synchronous methods under random times and adversarial partial participation of workers, and prove that their time complexities are optimal in many practical regimes, up to logarithmic factors. While synchronous methods are not universal solutions and there exist tasks where asynchronous methods may be necessary, we show that they are sufficient for many modern heterogeneous computation and communication scenarios.
Show more
C3P: Contrastive promoter-protein pretraining yields representations capturing bacterial gene regulation
Cameron Dufault ⋅ Scott Xu ⋅ Alan Moses
Despite the increasing scale of genome language models (gLMs), their ability to decode the function of regulatory sequences remains unclear. gLM pretraining relies on sequence reconstruction, which may struggle due to the noisy, rapidly evolving nature of regulatory DNA. Self-supervised contrastive approaches provide a promising alternative. Inspired by language-image architectures like CLIP, we demonstrate contrastive promoter-protein pretraining (C3P). By learning to align promoters to their corresponding proteins, we leverage the rich representations of proteins learned by protein language models as supervisory signal for the learning of promoter representations. After training on 88 million bacterial promoter-protein pairs, we evaluate the predictive power of C3P-learned promoter representations for inference of curated regulatory annotations, finding multi-fold improvement over leading gLMs. We also introduce zero-shot co-regulated gene retrieval, the ability to find co-regulated genes in a genome using no experimental data. We find that compared to a randomly initialized baseline, C3P training consistently provides significant zero-shot performance gains, unlike gLMs. Scaling analysis reveals the potential for further improvement as well as the efficiency of C3P, which achieved strong performance at a fraction of the training cost of leading gLMs. In addition to demonstrating that C3P training is effective for learning representations of bacterial regulatory sequences, our strong zero-shot co-regulated gene retrieval performance suggests the possibility of decoding gene regulation for millions of bacteria from their genomes alone.
Show more
Agent Security is a Systems Problem
Mihai Christodorescu ⋅ Earlence Fernandes ⋅ Ashish Hooda ⋅ Somesh Jha ⋅ Johann Rehberger ⋅ Kamalika Chaudhuri ⋅ Xiaohan Fu ⋅ Khawaja Shams ⋅ Guy Amir ⋅ Jihye Choi ⋅ Sarthak Choudhary ⋅ Nils Palumbo ⋅ Andrey Labunets ⋅ Nishit V Pandya
We take the position that agent security must be approached as a systems problem: the AI model powering the agent must be treated as an untrusted component, and security invariants must be enforced at the system level. Through this lens, efforts to increase model robustness (the dominant viewpoint in the community) are insufficient on their own. Instead, we must complement existing efforts with techniques from the systems security domain. Based on our experience as cybersecurity researchers in operating systems, networks, formal methods, and adversarial machine learning, we articulate a set of core principles, grounded in decades of systems security research, that provide a foundation for designing agentic systems with predictable guarantees. As evidence, we analyze eleven representative real-world attacks on agents and discuss how systems principles, if realized, could have prevented these attacks. We also identify the research challenges that stand in the way of implementing these principles in agents.
Show more
We Should Distinguish Unlearning From Untraining
Eleni Triantafillou ⋅ Imtiaz Humayun ⋅ Mónica Ribero ⋅ Alexander Turner ⋅ Michael Mozer ⋅ Georgios Kaissis
There has been a recent surge in interest about the question of how we can "delete" specific data points or behaviours from a trained model, a goal referred to as "machine unlearning". We argue that the umbrella term "unlearning" actually spans two distinct problem formulations, but the distinction between them has not yet been observed in literature. This causes ambiguity around when an unlearning algorithm is expected to work, leads to the use of inappropriate metrics and baselines when comparing algorithms to one another, difficulty in interpreting results, and missed opportunities for pursuing critical research directions. In this paper, we argue for the position that a fundamental distinction must be made between two notions that we refer to as Unlearning and Untraining. On one hand, Untraining aims to reverse the effect of having trained on a given "forget set", i.e. to remove the influence that that specific set of examples had on the model during training. On the other hand, Unlearning aims not just to remove the influence of those given examples, but also of the entire underlying subdistribution from which they were sampled (and thus e.g. the concept or model behaviour that those examples represent). We discuss technical definitions of these problems and map problem settings studied in the literature to each notion. By disambiguating technical definitions, our work aims to accelerate progress in this important field.
Show more
AdaSRU: Adaptive Source-Free Recommendation Unlearning via Gradient-Constrained Optimization
Xiaoran Zhao ⋅ Weiming Liu ⋅ Lianyong Qi ⋅ Weiyi Zhong ⋅ Yuwen Liu ⋅ Xiaolong Xu ⋅ Haolong Xiang ⋅ Qiang Ni ⋅ Xuyun Zhang ⋅ Wanchun Dou ⋅ Xiaokang Zhou
Recommendation unlearning (RU) aims to remove the influence of specified user--item interactions from a trained recommender while preserving utility on the remaining data. Existing RU methods usually require access to the original training data, especially the remaining data, to recover recommendation utility after unlearning. However, access to such data is often restricted or unavailable in practice due to privacy constraints, storage costs, or third-party deployment, giving rise to source-free recommendation unlearning. To address this challenge, we propose an adaptive source-free recommendation unlearning (**AdaSRU**) method that enables effective unlearning without accessing the original remaining data. **AdaSRU** first estimates the inaccessible gradient on the remaining data. It exploits the first-order optimality condition of the trained recommender and approximates the remaining gradient using the forgetting data and squared gradient accumulators from adaptive optimizers, e.g., Adam. **AdaSRU** then formulates unlearning as a gradient-constrained update, where the estimated remaining gradient is used as a constraint to limit utility degradation while maximizing the unlearning objective. **AdaSRU** adaptively adjusts the utility-preserving correction according to the current gradient geometry. We further theoretically prove that **AdaSRU** approaches an approximate Pareto stationary solution. Extensive experiments on three real-world datasets demonstrate that **AdaSRU** achieves strong performance in both unlearning and recommendation. The code is available at https://anonymous.4open.science/r/AdaSRU-94D9/.
Show more
Beyond Family Labels: A Taxonomic Ornstein-Uhlenbeck Prior for Avian 3D Shape Recovery
Yunchan Jeon ⋅ SooGon Kim ⋅ Seungwook Kim ⋅ CheolWon Lee ⋅ Jongmin Lee
Recovering 3D pose and shape of birds from single images is bottlenecked by the long tail of avian diversity: out of $\sim 10{,}000$ species, only a few dozen have 3D-labeled training data, and in-the-wild test sets routinely contain species never seen at training. Generalizing across clades is therefore a load-bearing problem. The current state of the art, AniMer+, mitigates it with a family-level contrastive loss that conveys only a binary same-family indicator per pair and discards the continuous structure of the species tree. We propose a **taxonomic Ornstein--Uhlenbeck prior** on the AVES shape code: a multivariate Gaussian whose covariance decays exponentially with Linnaean rank distance, with a learnable selection strength and noise scale. The prior generalizes family contrastive into a continuous, generative form, exposing the graded cross- and within-family similarity that a binary indicator cannot, and brings continuous comparative phylogenetics in as a biology-grounded inductive bias for the long tail. Our method consistently improves over AniMer+ on every reported avian benchmark cell, with the largest gains on the unseen-species CowBird benchmark. A permutation ablation isolates taxonomic structure from generic Gaussian smoothing, and a rank-granularity ablation identifies cross-order distinctions—the structure beyond a family-vs-not indicator—as the dominant source of the unseen-species gain.
Show more
TALK: One-Shot Batch Design for Protein Variant Effect Prediction
qiao huang ⋅ Ao Shen ⋅ Jie Du ⋅ Manning Wang
Protein variant effect prediction is central to understanding protein function and guiding protein engineering, yet experimental labels remain costly. In many protein experiments, the variants that can be assayed are dictated by protocol constraints, while the variants that need accurate prediction may lie in a different or broader region of the landscape. This assay-prediction mismatch makes candidate-local criteria such as predicted function, uncertainty, or exploration value insufficient for measurement design. Given a protocol-defined accessible space and a limited budget, we ask which variants should be assayed so that the resulting data best improves landscape prediction. We propose TALK, which estimates each candidate’s contribution through posterior coupling between accessible and target variants and reduces redundant information within the batch. Across ProteinGym and GB1 settings spanning mutation-order transfer and different accessible-target relationships, TALK uses limited budgets more efficiently, yielding labels that better support subsequent prediction than baselines. These results point to a practical two-stage workflow: first, spend a limited assay budget on variants selected to be informative for the target landscape and train a predictive model from the measured labels; then use this model to score and prioritize high-potential variants.
Show more
Nemotron-Cascade: Scaling Cascaded Reinforcement Learning for General-Purpose Reasoning Models
Boxin Wang ⋅ Chankyu Lee ⋅ Nayeon Lee ⋅ Sheng-Chieh Lin ⋅ Wenliang Dai ⋅ Yang Chen ⋅ Yangyi Chen ⋅ Zhuolin Yang ⋅ Zihan Liu ⋅ Mohammad Shoeybi ⋅ Bryan Catanzaro ⋅ Wei Ping
Building general-purpose reasoning models with reinforcement learning (RL) entails substantial cross-domain heterogeneity, including large variation in inference-time response lengths and verification latency. Such variability complicates the RL infrastructure, slows training, and makes training curriculum (e.g., response length extension) and hyperparameter selection challenging. In this work, we propose cascaded domain-wise reinforcement learning (Cascade RL) to develop Nemotron-Cascade, capable of operating in both \emph{instruct} and deep \emph{thinking} modes, without any performance gap relative to a thinking-only counterpart. Departing from conventional approaches that blend heterogeneous prompts from different domains, Cascade RL orchestrates sequential, domain-wise RL, reducing engineering complexity and delivering state-of-the-art performance across a wide range of benchmarks. Notably, RLHF for alignment, when used as a pre-step, boosts the model's reasoning ability far beyond mere preference optimization, and subsequent domain-wise RLVR stages rarely degrade the benchmark performance attained in earlier domains and may even improve it. Our 14B model, after RL, outperforms its SFT teacher, DeepSeek-R1-0528, on LiveCodeBench v5/v6/Pro and achieves silver-medal performance in the 2025 International Olympiad in Informatics (IOI). We transparently share our training and data recipes.
Show more
Reference-Guided Training: Adaptive Gradient Scaling via Per-Sample Loss Comparisons
Israel Rodrigues Soares ⋅ Pedro Benedetti ⋅ Thanda Shwe ⋅ Israel Mendonca ⋅ Masayoshi Aritsugi
Empirical Risk Minimization treats training samples uniformly, which can degrade performance in noisy or heterogeneous settings. We introduce a reference-guided objective that uses per-sample error comparisons with a fixed reference model to reweight gradient contributions without relying on prediction imitation, while preserving gradient direction and ensuring bounded scaling. Theoretical analysis shows that the formulation induces adaptive weighting with controlled amplification. Experiments on image classification with synthetic label noise and on time-series forecasting demonstrate broad improvements across architectures, particularly in moderate-to-high noise and heterogeneous regimes. Results further indicate that the method is most effective when the error-ranking alignment between the baseline model and the reference model is low, suggesting that gains arise primarily from the diversity of the reference error ranking relative to the learner, rather than the reference's absolute performance.
Show more
CLIOPATRA: Extracting Private Information from LLM Insights
Meenatchi Sundaram Muthu Selva Annamalai ⋅ Emiliano De Cristofaro ⋅ Peter Kairouz
The widespread adoption of AI assistants has prompted the development of privacy-aware platforms designed to extract insights from real-world usage. Their privacy protections primarily rely on layering multiple heuristic techniques, such as PII redaction, clustering, and aggregation. In this paper, we put their privacy claims to the test by presenting CLIOPATRA, the first attack against privacy-preserving LLM-based insights systems. Our attack involves an adversary that carefully designs and inserts malicious chats into the system to break multiple layers of protections and induce the leakage of sensitive information from a target user’s chat. We evaluate CLIOPATRA on one such platform, Anthropic’s Clio, and target synthetically generated medical chats to show that an adversary can successfully and confidently (with nearly 100% precision) extract the medical history contained in these chats in up to 65% of cases. We also show that CLIOPATRA can stealthily extract information by obfuscating the private information in the generated insights. Finally, we demonstrate that existing ad hoc mitigations, such as LLM-based privacy auditing, are unreliable and fail to detect major leaks. Taken together, our findings indicate that, even when layered, current heuristic protections are insufficient to adequately protect user data, and that prompt injection has been an understudied risk in LLM-based insight systems.
Show more
Felid: A Flexible and Efficient Design for Transformer Fine-Tuning over Encrypted Data
Linru Zhang ⋅ Jun Jie Sim ⋅ Xiangning Wang ⋅ Jiahao Zhong ⋅ Kaiyu Zhou ⋅ Huanyi Ye ⋅ Yongsen Zheng ⋅ Lushan Song ⋅ Xiaojian Liang ⋅ Yingting Liu ⋅ Yujing Sun ⋅ Huaxiong Wang ⋅ Pu Duan ⋅ Kwok-Yan Lam
Fully Homomorphic Encryption (FHE) enables non-interactive, privacy-preserving transformer fine-tuning, protecting highly sensitive data as inputs to large language models. Prior work has predominantly pursued efficiency gains through architectural refinements, yet overall performance remains bounded by the substantial cost inherent in encrypted computation. To address these issues, we target the bottleneck at the ciphertext operation level. Concretely, we propose a novel packing strategy that supports batching over a flexible number of inputs, making it well-suited for fine-tuning workloads. In addition, we design operation-specific matrix evaluation algorithms that significantly reduce (i) the number of ciphertext rotations and relinearizations, and (ii) the number of bootstrapping invocations. Additionally, we exploit functional bootstrapping to fold the evaluation of element-wise activation functions directly into the bootstrapping step, thereby reducing overall computational complexity. As a result, our framework achieves a $30\times$ reduction in ciphertext rotations and relinearizations across all matrix operations, as well as a $2.5\times$ reduction in FHE bootstrapping invocations over the entire computation flow, compared to the latest state-of-the-art work. It is worth noting that our system completes fine-tuning of a 2-layer BERT-style model with a batch size of 16 in 650 seconds, achieving a $8\times$ speedup over the state-of-the-art work.
Show more
Compact SO(3) Equivariant Atomistic Foundation Models via Structural Pruning
Chen Wang ⋅ Siyu Hu ⋅ Guangming Tan ⋅ Weile Jia
SO(3) equivariant graph neural networks have become the dominant paradigm for atomistic foundation models, achieving high accuracy and data efficiency by building rotational symmetry directly into the architecture. Yet the computational cost of their higher-order tensor operations creates a tough trade-off between model accuracy and inference efficiency. In this paper, we propose a structural pruning method for SO(3) equivariant atomistic foundation models to bridge this accuracy-efficiency gap. The pruning is applied along the channel and order dimensions, with each irreducible representation kept or removed as a complete block, thereby retaining SO(3) equivariance. Starting from a large checkpoint, the pruned model substantially reduces the inference cost while retaining higher accuracy than an independently trained small model. The pruned MACE-MP model outperforms the official from-scratch trained small model on 7 of 9 metrics on the Matbench Discovery leaderboard. In terms of efficiency, compressed MACE-MP and MACE-OFF models contain 1.5$\times$ to 4$\times$ fewer parameters and require 2.5$\times$ to 4$\times$ less pre-training compute than training a small model from scratch. For downstream applications, fine-tuning the pruned model reduces energy and force errors by 70.1\% and 34.4\% compared to training task-specific models from scratch across eight representative downstream datasets. We demonstrate that the method generalizes to other SO(3) equivariant architectures (SevenNet, eSCN) and can be combined with quantization and knowledge distillation for further gains.
Show more
Follow the Regularized Leader Does Not Converge in Constrained Optimization
Ioannis Anagnostides ⋅ Ioannis Panageas ⋅ Nikolas Patris ⋅ Tuomas Sandholm
Follow the regularized leader (FTRL) is a foundational algorithm in online learning whose regret properties have been extensively studied for decades. However, in stark contrast to mirror descent---and gradient descent in particular---the convergence of FTRL to first-order stationary points in constrained nonconvex optimization was hitherto unresolved. In this paper, we show that continuous-time FTRL can fail to converge even asymptotically in constrained optimization problems; that is, the Karush-Kuhn-Tucker (KKT) gap of FTRL can remain bounded away from zero indefinitely. This is especially surprising in light of the fact that continuous-time FTRL guarantees a monotonic improvement of the objective value. As a result, we establish that non-convergent behavior of no-regret dynamics---which besets general variational inequality problems---can occur even in potential systems. From a technical standpoint, our construction relies on an infinitely differentiable function where the gradient flow dynamics exhibit decelerating oscillations without ever converging pointwise, which we in turn embed into higher-dimensional FTRL dynamics that sustain a perpetually large KKT gap.
Show more
CURE: Counterfactual Unsafe-token Re-masking for Diffusion Large Language Model Test-time Alignment
Lichao Wang ⋅ Mingquan Zhang ⋅ Xian Zhang ⋅ Juntao Dai ⋅ Chi Liu ⋅ Yaodong Yang ⋅ Jingwei Yi
Diffusion language models (DLMs) generate text through iterative denoising, enabling parallel decoding, bidirectional conditioning, and editable intermediate states. Under harmful prompts, however, intermediate denoising states can contain risk-inducing tokens that are seemingly benign but increase the probability of an unsafe final response by shaping how the remaining masked positions are completed. Since committed tokens in DLMs can still be re-masked and rewritten, safety control can be applied during denoising by removing a small set of risk-inducing tokens before they drive subsequent generation toward unsafe responses. Existing DLM defenses mainly rely on model-level alignment, trajectory-level detection, or block-level repair, often requiring training cost, extra inference passes, or coarse regeneration. We propose CURE, a counterfactual value-driven test-time alignment method that keeps the base DLM fixed and selectively re-masks risk-inducing tokens. CURE trains a time-conditioned safety value model to estimate whether a partially denoised state will lead to an unsafe final response. During inference, CURE constructs counterfactual readout views that remove or isolate targeted tokens, estimates their contribution to future unsafe response, and re-masks only high-risk tokens for later rewriting. Across three DLMs and five jailbreak benchmarks, CURE reduces macro-average ASR from 47.20\% to 1.36\%, preserves utility on MMLU and GSM8K, and adds only 0.84 extra TFLOPs, achieving a strong safety-utility-efficiency trade-off. Code is available at \url{https://anonymous.4open.science/r/CURE}.
Show more
Quantized Reasoning Models Think They Need to Think Longer, but They Do Not
Sanae Lotfi ⋅ Polina Kirichenko ⋅ Steven Li ⋅ Zechun Liu
Post-training quantization (PTQ) is widely used to deploy large language models efficiently, but its effect on reasoning models is not well understood. Across math, coding, and science QA, we find that aggressive PTQ reduces accuracy while increasing chain-of-thought (CoT) length. Surprisingly, we show that in up to 52% of the quantized models' failures, models reach the right answer in intermediate reasoning steps but do not output it as a final answer. To understand why quantization leads to this increase in overthinking errors, we measure the token-level KL divergence between quantized and full-precision output distributions. Positions with high KL divergence correlate strongly with high next-token entropy, and at these positions quantized models disproportionately sample overthinking markers such as “wait", “but", and “alternatively". We show that simply introducing a training-free logit penalty on a curated set of overthinking markers can reduce CoT length by 12--23% while preserving or improving accuracy across 5 models (1.5B--32B parameters), 3 quantization methods, and 5 benchmarks, yielding the best Pareto frontier of accuracy against reasoning cost. Overthinking errors produced by quantized models are particularly reduced by up to 58%.
Show more
BEAST3D: animal behavioral analysis and neural encoding from multi-view video via Gaussian splatting
Yanchen Wang ⋅ Lenny Aharon ⋅ Wangshu Zhu ⋅ Kyle Daruwalla ⋅ Linghua Zhang ⋅ Jiaru Zou ⋅ Selmaan N Chettih ⋅ X Helen Hou ⋅ Liam Paninski ⋅ Matthew Whiteway
Multi-view video recordings are increasingly used to capture the 3D movements of animals in experimental settings, yet extracting rich 3D representations from these recordings remains challenging. Supervised pose estimation requires extensive manual annotation, while general-purpose 3D reconstruction models trained on generic scene datasets fail on the specialized imagery and sparse-view setting of laboratory experiments. We address these limitations with BEAST3D, a self-supervised pretraining framework that learns 3D visual representations from unlabeled, calibrated multi-view video. BEAST3D uses a vision transformer to predict 3D Gaussian splats that reconstruct held-out views through differentiable rendering, while simultaneously segmenting the animal from the background. BEAST3D reconstructs 3D structure with as few as four views by conditioning directly on known camera parameters---unlike general-purpose models, which must estimate camera geometry from dense overlapping viewpoints that are seldom available in lab settings. Through comprehensive evaluation across four species, we demonstrate that BEAST3D produces rich, viewpoint-invariant features that transfer effectively to three downstream tasks: novel view synthesis, which validates the quality of the learned 3D representations; multi-view pose estimation, which provides the sparse keypoint trajectories widely used in behavioral analysis; and neural encoding, which relates 3D behavioral features to simultaneously recorded neural activity. BEAST3D thus establishes a versatile framework for behavioral analysis that leverages 3D structure in modern multi-view laboratory recordings.
Show more
CausalBind: Causal Modeling and Learning for Protein-Molecule Virtual Screening
Loka Li ⋅ Jin Tian ⋅ Kun Zhang
Protein-molecule virtual screening is increasingly cast as a problem of representation learning in a shared embedding space. Existing methods rely on dense holistic alignment, entangling invariant binding determinants with nuisance correlations and limiting transfer to new targets. It has been noted that binding in protein–molecule systems involves sparse cross-modality interactions: binding is governed by a small contact interface and a few decisive local interactions (e.g., hydrogen bonds, hydrophobic contacts, and salt bridges) rather than the global structures of the protein and molecule. We hypothesize that uncovering and leveraging sparse interaction patterns is critical for generalization beyond the training data, as these patterns are reusable and expected to improve performance across different scenarios. In this paper, we aim to identify and leverage sparse interaction patterns, and verify our hypothesis. Since the training data contain only observed binding pairs, we formalize this prior via a V-structure causal model under Heckman-style selection, and establish three theoretical results: (i) the latent concepts of interacting proteins and molecules are not identifiable without appropriate sparsity constraints;(ii) these concepts and their sparse interactions are component-wise identifiable under structural sparsity conditions; and (iii) a low-rank relaxation of these conditions yields subspace identifiability of the concepts and interactions. Inspired by these principles, we propose CausalBind with three implementation variants. Extensive experiments on DUD-E and LIT-PCBA benchmarks show that all variants consistently outperform strong baselines across all metrics, respectively.
Show more
From Average Sensitivity to Small-Loss Regret Bounds under Random-Order Model
Shinsaku Sakaue ⋅ Yuichi Yoshida
We study online learning in the random-order model, where the multiset of loss functions is chosen adversarially but revealed in a uniformly random order. By extending the *batch-to-online* transformation of Dong and Yoshida (2023), we show that if an offline algorithm enjoys a $(1+\varepsilon)$-approximation guarantee, an *average sensitivity* bound controlled by a function $\varphi(\varepsilon)$, and stability with respect to $\varepsilon$, then we can obtain a *small-loss* regret bound typically of order $\tilde O(\varphi^{\star}(\mathrm{OPT}_T))$, where $\varphi^{\star}$ is the concave conjugate of $\varphi$, $\mathrm{OPT}_T$ is the offline optimum over $T$ rounds, and $\tilde O$ hides polylogarithmic factors in $T$. Our result refines their original $(1+\varepsilon)$-approximate regret guarantee and applies to a broad class of problems, including online $k$-means clustering and online low-rank approximation. We further apply our approach to online submodular function minimization using $(1\pm\varepsilon)$-cut sparsifiers of submodular hypergraphs, obtaining a small-loss regret bound of $\tilde O(n^3 + n^{3/4}\mathrm{OPT}_T^{3/4})$, where $n$ is the ground-set size; we also demonstrate its applicability to online $\ell_1$ regression. Our work sheds light on the power of sparsification and related algorithmic techniques in achieving small-loss regret bounds in the random-order model, without requiring structural assumptions on loss functions, such as linearity or smoothness.
Show more
Déjà View: Looping Transformers for Multi-View 3D Reconstruction
Alessandro Burzio ⋅ Tobias Fischer ⋅ Sven Elflein ⋅ Qunjie Zhou ⋅ Riccardo de Lutio ⋅ Jiawei Ren ⋅ Jiahui Huang ⋅ Shengyu Huang ⋅ Marc Pollefeys ⋅ Laura Leal-Taixé ⋅ Zan Gojcic ⋅ Haithem Turki
Recent feed-forward 3D reconstruction transformers have scaled to over a billion parameters, following the broader trend of increasing model capacity in computer vision. Yet emerging evidence suggests that contiguous transformer layers often behave like repeated applications of similar operations, and multi-view reconstruction transformers refine their predictions progressively across decoder depth. We posit that model depth partially buys iteration, paid for inefficiently in unique parameters, and instead make that iteration explicit in architecture. Our model, DéjàView, applies a single looped transformer block recurrently to per-view features for K refinement steps. Trained once, it exposes K as an inference-time compute knob, matching or outperforming substantially larger feed-forward baselines across five reconstruction benchmarks spanning indoor, outdoor, object-centric, and driving scenes, while using a fraction of their parameters and comparable or lower compute. Importantly, the same looped block formulation outperforms an otherwise identical variant with independent per-step parameters under matched training data and compute, suggesting that explicit iteration is not merely a compute-efficient substitute for capacity but a stronger inductive bias for multi-view 3D reconstruction.
Show more
The Attacker in the Mirror: Breaking Self-Consistency in Safety via Anchored Bipolicy Self-Play
Gabriele La Malfa ⋅ Emanuele La Malfa ⋅ Saar Cohen ⋅ Jie Zhang ⋅ Michael Luck ⋅ Michael Wooldridge ⋅ Elizabeth Black
Self-play red team is an established approach to improving AI safety in which different instances of the same model play attacker and defender roles in a zero-sum game, i.e., where the attacker tries to jailbreak the defender; if self-play converges to a Nash equilibrium, the model is guaranteed to respond safely within the settings of the game. Although the parameter sharing enforced by the use of the same model for the two roles improves stability and performance, it introduces fundamental theoretical and architectural limitations. We show that the set of Nash equilibria that can be reached corresponds to a broad class of behaviours that includes trivial always refuse strategies and oracle-like defenders, thus limiting practical applicability. We then show that when attacker and defender share and update the same base model, the dynamics collapse to self-consistency, so that attacks do not enforce adversarial pressure on the defender. In response, we propose Anchored Bipolicy Self-Play, which trains distinct role-specific LoRA adapters on top of a frozen base model, thereby maintaining stable optimisation while preserving adversarial pressure through explicit role separation. In relation to standard self-play, we show up to 100x greater parameter efficiency than fine-tuning and consistent improvements in safety compared to self-play fine-tuned models. We evaluate on Qwen2.5-\{3B, 7B,14B\}-IT models across widely used safety benchmarks, showing improved robustness without loss of reasoning ability. Cross-play experiments further show that our attacker and defender models are superior to self-play in terms of adversarial defence and safety.
Show more
Pruning and Distilling Mixture-of-Experts into Dense Language Models
Junhyuck Kim ⋅ Jihun Yun ⋅ Haechan Kim ⋅ Gyeongman Kim ⋅ Joonghyun Bae ⋅ Jaewoong Cho
Mixture-of-Experts (MoE) is now the dominant architecture for frontier language models, yet it requires all expert parameters to be loaded in memory, making it less preferable for memory-constrained deployment. Existing compression methods reduce the number of experts but the output remains an MoE model with the same fundamental limitation. We present the first systematic framework for converting a trained MoE into a standard fully dense architecture: experts are scored, selected, and grouped, then concatenated into a dense FFN and refined by knowledge distillation from the MoE teacher. We evaluate 7 scoring, 5 grouping, and 2 magnitude scaling methods across a range of selected expert counts on Qwen3-30B-A3B, yielding 350 configurations. We find that the choice of scoring method is the most impactful, with our novel diversity-aware scoring consistently outperforming prior methods on Qwen3-30B-A3B, DeepSeek-V2-Lite, and GPT-OSS-20B. Under a controlled comparison at matched parameter count, MoE-to-dense outperforms dense-to-dense pruning by $+$6.3 pp in average downstream accuracy after ~4B-token distillation at 1.6x faster training wall-clock speed.
Show more
Catch-Only-One: Non-Transferable Examples for Model-Specific Authorization
Zihan Wang ⋅ Ethan Ma ⋅ Zhongkui Ma ⋅ Shuofeng Liu ⋅ Akide Liu ⋅ Derui Wang ⋅ Minhui Xue ⋅ Guangdong Bai
Recent AI regulations increasingly emphasize the need for mechanisms that preserve the utility of data for AI innovation while preventing misuse, particularly by enforcing purpose limitation in downstream AI applications. In practice, enforcing this principle remains challenging, as released data can be trivially fed into arbitrary models beyond its declared intent. Existing approaches attempt to mitigate this risk by either perturbing data or retraining models to limit unintended use. These strategies, however, offer no protection against inference by unknown or externally trained models, or fundamentally rely on control over the training or deployment. In this work, we introduce non-transferable examples (NTEs), recoded data that act as a task-level "ciphertext" decodable only by a designated model. Where adversarial examples exploit sensitive input directions, NTEs use the complementary insensitive subspace: a training-free, data-agnostic recoding within a model-specific low-sensitivity subspace preserves the authorized model's outputs while degrading unauthorized ones through subspace misalignment. We establish formal bounds certifying output fidelity for the authorized model and showing unauthorized degradation scales with measurable spectral misalignment between models. Empirically, NTEs preserve authorized-model performance across diverse vision backbones and vision-language models, while unauthorized models collapse even under adaptive reconstruction attacks. These results establish NTEs as a practical means to preserve intended data utility while preventing unauthorized exploitation. Our source code and visual demos are available at: https://github.com/model-specific/non-transferable-examples.
Show more
Testable and Actionable Calibration for Full Swap Regret
Konstantina Bairaktari ⋅ Lunjia Hu ⋅ Huy Nguyen ⋅ Jonathan Ullman
AI generated predictions increasingly inform decision making in critical tasks, and therefore must be trustworthy. One widely used measure of trustworthiness is calibration, which requires that the predictions match the true frequencies and can be treated like real probabilities of a given outcome. However, defining calibration is subtle, and designing good measures of calibration error has been an active topic of recent research. The first goal is to find calibration measures that are *actionable*, meaning they can inform decision makers about their utility loss when predictions are treated as true probabilities, which is known as swap regret. The second goal is to find calibration measures that are *testable*, meaning that calibration error can be measured from a small sample of predictions and outcomes. Although these are very basic requirements, there is no existing calibration measure that fully satisfies both properties, and all existing measures relax actionability by bounding a weaker notion of swap regret, or relax testability by having suboptimal estimation error. We introduce a new calibration measure, Soft-Binned Calibration Decision Loss (SCDL), which we prove is fully actionable without weakening either requirement, and testable with nearly optimal error rate. In addition, SCDL satisfies other desired properties such as continuity and consistency. We also provide a set of experiments confirming that the theoretical advantages of SCDL compared to other measures lead to better performance in practice.
Show more
LittleLearner: Language Models Under Pedagogically-Controlled Knowledge Exposure
Fanfei Li ⋅ Jana Zeller ⋅ Manuel Prada-Corral ⋅ Thaddäus Wiedemer ⋅ Prasanna Mayilvahanan ⋅ Ryan Cotterell ⋅ Wieland Brendel
Modern language models are trained on heterogeneous web-scale text corpora. Consequently, studying knowledge and skill acquisition is difficult, as prior exposure to related content is hard to characterize. To address this challenge, we introduce **LittleCurriculum**, a meticulously curated 88B-token pretraining corpus tailored to U.S. elementary school material, explicitly excluding concepts, facts, and vocabulary taught above Grade 5. Training a 2B-parameter LLM from scratch on **LittleCurriculum** yields **LittleLearner**, a model with sufficient language competence for open-ended evaluation, yet with clear knowledge and capability boundaries mapped to interpretable curriculum guidelines. We release **LittleCurriculum** and **LittleLearner** as a developmentally restricted sandbox to study how models acquire, represent, and use data under a well-defined training scope. We illustrate the sandbox's utility in a first suite of experiments on injecting new knowledge through post-training and in-context learning. These methods let **LittleLearner** better utilize existing knowledge, but do not raise out-of-scope capabilities. Our findings underscore the value of this controlled environment for future investigations.
Show more
Can We Model the Artifacts Explicitly? Disentangle Artifacts via Pairwise Edit Relations for Image Manipulation Localization
Xuekang Zhu ⋅ Kaiwen Feng ⋅ Ruifeng Wang ⋅ Xiwen Wang ⋅ Xiaochen Ma ⋅ Bo Du ⋅ Changjiang Jiang ⋅ Chenfan Qu ⋅ Songyu Ye ⋅ Jian liu ⋅ Ji-Zhe, Knight Zhou
Image Manipulation Localization (IML) is commonly formulated as a fully supervised learning task that estimates the optimal manipulation mask $y$ for a given image $x$. In this work, we first reveal the latent nature of artifacts and thus reinterpret IML as a latent-variable problem, $P(y| x)=\int P(y|z) P(z|x)dz$, where $z$ denotes the artifacts. Following this interpretation, we pinpoint the cause for the current IML models' insufficiency as their implicit artifacts modeling strategy, highlighting the necessity of modeling $z$ in an explicit manner. Without direct labels, feature disentanglement is the most appropriate solution for this explicit modeling. Accordingly, we propose a two-stage learning paradigm with the Pairwise Artifacts Learning (PAL) and Standard Localization (SL) phases to estimate $P(z|x)$ and $P (y|z)$ via edit relations. To support our edit-relation-based learning, we further curate EditGroup-45K, a source-anchored dataset organized into edit groups for pair construction. Extensive experiments show that our PAL paradigm yields consistent improvements across diverse IML architectures, and empirical analyses further verify that PAL does capture artifacts explicitly through feature disentanglement. Code and dataset will be publicly available.
Show more
Emergent Biological Capabilities in a Foundation Model for Molecular Interactions
Adi Mashiach
Foundation models trained on large unlabeled corpora develop emergent capabilities — tasks that go beyond the training objective. We extend this paradigm to multimodal biology with Bonbon, a foundation model for molecular interactions trained on approximately 3.2 trillion tokens of paired protein-ligand sequences. From a single self-supervised objective, Bonbon exhibits four zero-shot emergent capabilities at hierarchical levels of resolution — functional (mechanism of action), residue (binding site localization, including orthosteric/allosteric discrimination), bond (covalent versus non-covalent engagement), and atom (active moiety identification at single-bond resolution). All four capabilities generalize to out-of-distribution proteins. Bonbon was trained on protein-ligand sequences alone, without any task-specific supervision and without structural coordinates. We conducted controlled scaling experiments across four model scales from 42M to 1.6B parameters; every capability evaluated shows a sharp jump at the largest scale on continuous metrics, even as pretraining loss decreases smoothly. This demonstrates that loss-based scaling laws alone are insufficient to predict when zero-shot emergent capabilities become available. The capabilities reported here emerge after training at a 1:2,000 parameter-to-token ratio to accommodate the sparsity of paired biological interaction data. The same frozen representations achieve state-of-the-art binding and affinity predictions at five orders of magnitude greater speed than structure-based methods. In prospective wet-lab validation, Bonbon achieves a 70% hit rate on de novo compounds, discovering novel chemotypes not previously documented as inhibitors for those targets. All of the tested compounds had less than 60% structural similarity to molecules in the training corpus.
Show more
Behavioral Geometric Supervision Aligns Video Foundation Models with Human Social Perception
Kathy Garcia ⋅ Leyla Isik
Current video foundation models, including the strongest self-supervised models such as V-JEPA2, fail to capture how humans organize social information in dynamic scenes. For example, across a range of diverse vision models tested, none were able to predict human similarity judgments to social video clips as well as a sentence embedding model of the caption text (MPNet). We show this gap in vision model performance can be closed by a compact behavioral supervisory signal. We introduce behavioral geometric supervision (BGS): a hybrid objective that constrains local and global pairwise embedding geometry to match the relational similarity structure across videos. We apply this method using a new human similarity dataset, containing 49,484 odd-one-out judgments from 250 naturalistic social video clips, and low-rank adaptation across four ViT backbones (V-JEPA 2/2.1, TimeSformer, VideoMAE, and CLIP). We find that one of the best fine-tuned models, V-JEPA 2.1, nearly triples in performance compared to the pre-trained baseline and reaches close to the noise ceiling, exceeding the strongest sentence-embedding baseline. In addition, finetuned models (i) capture unique variance in human judgments that caption-based language embeddings do not, (ii) develop interpretable social-affective attributes (valence, arousal, and dominance) despite never being trained on any of these attributes, (iii) zero-shot transfer to a separate dataset of out-of-distribution abstract social interactions, and (iv) shift spatial attention from scene context to socially informative regions (faces, gaze, and interacting bodies). A matched language-distillation control fails to reproduce these gains, ruling out caption transfer as the mechanism. Our results show how a modest amount of human behavioral data can steer video models toward human-like social visual understanding.
Show more
Learning to Audit ML Models with Theory of Mind
Yao Rong ⋅ Vaibhav Unhelkar
When auditing machine learning models, human experts sequentially inspect model predictions and explanations, gradually building a mental model of the model’s behavior to uncover errors. This manual process is critical for safe deployment in high-stakes domains but fundamentally limited in scalability. In this work, we propose a novel method to simulate this human auditing behavior for automating the auditing process. We formalize auditing as a sequential decision-making problem, modeled as a Markov decision process. Our state representation summarizes the auditor’s accumulated knowledge of model predictions and feature usage. Building on this formalization, we solve the MDP using reinforcement learning, yielding RLAuditor, which learns to select informative test samples to efficiently uncover model errors. We further connect our formalization to Theory of Mind, showing that our state mirrors how human auditors build mental models: its update rule is equivalent to Bayesian belief revision and it converges to the model’s behavioral pattern at an interpretable rate. We validate our approach on multiple ML models and datasets across different data modalities. In a human subject study, we demonstrate that RLAuditor helps human auditors produce more accurate audit reports in less time.
Show more
Prediction Under Imperfect Compression: A Theory of Approximate MDL
Qian Li ⋅ Xinyu Mao ⋅ Shang-Hua Teng ⋅ Guangxu Yang
Minimum Description Length (MDL) formalizes the principle of Occam's razor by optimizing the total description length: $L(\mathrm{model})+L(\mathrm{data} \ | \ \mathrm{model})$. For sequential prediction, the MDL method repeatedly selects a model with a minimum objective score of the observed prefix for the next step prediction. Classical MDL prediction theory shows that exact optimization of the MDL objective indeed provides a strong compression guarantee that supports reliable prediction. However, practical machine learning usually can only find models by approximately optimizing the objective function. To bridge this gap, this paper addresses the following fundamental question: \emph{Under what forms of approximation and regularization does approximate MDL still guarantee reliable sequential prediction?} This work offers a principled characterization. We prove that for any approximation with \emph{additive} slack \(C\) of the more general form of the \emph{balanced} MDL objective: $\lambda\cdot L(\mathrm{model})+L(\mathrm{data} \ | \ \mathrm{model})$, the cumulative expected squared prediction error is finite for all $\lambda\ge1$. The case $\lambda>1$ is proved by an affinity-telescoping argument, while the boundary case $\lambda=1$ is proved by a likelihood-ratio stopping argument based on exact static MDL bounds. Our results establish that classical MDL regularization remains robust to any fixed additive optimization error. Furthermore, we establish that our characterization of the approximate MDL framework is sharp: When $0<\lambda<1$, overfits can happen to incur infinite cumulative expected error in the universal class of estimable measures, and hence a strong form of model-complexity regularization is necessary. In addition, model selection may fail in every regularized regime $\lambda >0$, under multiplicative approximation, and thus, additive approximation is both sufficient and essential.
Show more
Strong Stochastic Flow Maps
Sam McCallum ⋅ Zander Blasingame ⋅ Timothy Herschell ⋅ Niklas Rindtorff ⋅ Alexander Tong ⋅ James Foster
Flow and diffusion models generate high-quality samples in many modalities; however, many network evaluations are required during inference due to numerical integration of an underlying differential equation. Flow maps alleviate this problem by learning the solution map of the differential equation directly, enabling few-step sampling. Yet, current methods are restricted to approximating the solution map of ODEs. These methods can be used to learn the transition kernel of an SDE, thereby obtaining a solution map that recovers the marginal distributions of the process (weak convergence) rather than the solution path (strong convergence). We propose Strong Stochastic Flow Maps (SSFMs) as a novel framework for learning the strong solution map of additive-noise SDEs, directly generalizing deterministic flow maps to the stochastic setting. A polynomial approximation to Brownian motion is introduced and shown to converge pathwise. These results enable a simulation-free training objective for the solution map of diffusion models. We demonstrate that SSFMs outperform previous flow map methods on image generation and enable few-step sampling of molecular systems.
Show more
Benchmark Health Index: A Systematic Framework for Benchmarking the Benchmarks of LLMs
Longyuan Zhu ⋅ Hairan Hua ⋅ Linlin Miao ⋅ Wei Wang ⋅ 必应 赵
Large Language Models (LLMs) are advancing rapidly, yet the benchmarks used to measure this progress are becoming increasingly unreliable. Score inflation and selective reporting have eroded the authority of standard benchmarks, leaving the community uncertain about which evaluation results remain trustworthy. We introduce the **Benchmark Health Index (BHI)**, a pure data-driven framework for auditing evaluation sets along three orthogonal and complementary axes: (1) **Capability Discrimination**, measuring how sharply a benchmark separates model performance beyond noise; (2) **Anti-Saturation**, estimating remaining headroom before ceiling effects erode resolution and thus the benchmark's expected longevity; and (3) **Impact**, measuring capability-weighted benchmark adoption across the model ecosystem. By distilling **158** validated benchmarks from the technical reports of **117** representative models released between **2025** and **April 2026**, we systematically characterize the evaluation landscape. BHI is the **first framework** to quantify benchmark health at a macro level, providing a well-grounded quantitative basis for benchmark selection and supporting the dynamic filtering and continuous updating of evaluation sets. Our comprehensive analysis not only identifies benchmarks that remain highly credible but also exposes pervasive structural issues. By establishing a rigorous foundation for selection and management, BHI elevates evaluation practice from heuristic-based judgment to a quantifiable and interpretable scientific process. Our code is available online at
Show more
Graph Cascades: Contagion-Based Mesoscopic Rewiring for Structure-Aware Graph Machine Learning
Meher Chaitanya Pindiprolu ⋅ My Le ⋅ Luana Ruiz
We introduce **Graph Cascades**, a mesoscopic rewiring strategy for Graph Neural Networks (GNNs) and Graph Transformers (GTs) that captures intermediate-scale graph structure beyond purely local edges or fully global attention. Using contagion-based diffusion processes, Graph Cascades constructs, in $\mathcal{O}(|V|+|E|)$ time, an auxiliary graph where node pairs supported by repeated multi-hop reinforcement are promoted to direct neighbors. We theoretically characterize when reinforcement-based rewiring helps: sufficient conditions under which a reinforcement-based edge-selection rule has higher label agreement than direct adjacency, an SBM witness in which two-hop reinforcement is perfectly homophilic, and formalizing mesoscopic connectivity via graph effective resistance. Empirically, across node-classification benchmarks, Graph Cascades improves multiple GNN and sparse-GT backbones, with the most reliable gains observed on heterophilic and moderate- to high-degree homophilic graphs. The theoretical conditions also identify regimes where mesoscopic rewiring is unlikely to be beneficial -- low-degree regular graphs and graphs with structural bottlenecks -- and these predictions match the observed failures. We additionally observe tight correlations between performance and structural properties in the rewired graphs.
Show more
TiRex-2: Generalizing TiRex to Multivariate Data and Streaming
Patrick Podest ⋅ Marco Pichler ⋅ Elias Bürger ⋅ Levente Zólyomi ⋅ Bernhard Voggenberger ⋅ Wilhelm F Berghammer ⋅ Daniel Klotz ⋅ Sebastian Böck ⋅ Günter Klambauer ⋅ Sepp Hochreiter
We introduce TiRex-2, a recurrent xLSTM-based time series foundation model that generalizes the univariate TiRex to multivariate forecasting with both past and future covariates. Real-world forecasting is inherently sequential: observations arrive continuously, variables evolve jointly, and a subset of covariates is known ahead of time. Existing Transformer-based time series foundation models capture cross-variate dependencies but incur quadratic complexity in context length and require full-history recomputation as new observations arrive. TiRex-2 addresses these limitations through a memory-centric recurrent design that operates at constant per-patch cost under streaming. The model combines a bidirectional time mixer with an asymmetric grouped-attention variate mixer, enabling the integration of future-known covariates while preserving strict causality over target variables. To our knowledge, this is the first time series foundation model that achieves this combination of properties. To support scalable multivariate pretraining, we propose a synthetic coupling pipeline that composes diverse multivariate samples on the fly from large univariate corpora. Empirically, TiRex-2 achieves state-of-the-art zero-shot performance on GIFT-Eval and fev-bench, remains stable when streamed to arbitrary context lengths, and maintains constant inference cost per patch. The model uses 38.4M active parameters in univariate mode, with an additional 44.1M parameters activated for multivariate forecasting.
Show more
The Aleatoric-Epistemic Dichotomy of Uncertainty is Meaningful and Indispensable for Machine Learning
Yusuf Sale ⋅ Nikita Kotelevskii ⋅ Maxim Panov ⋅ Eyke Hüllermeier
The quantification of uncertainty, in particular the distinction between aleatoric and epistemic uncertainty, is receiving growing interest in machine learning. However, both its conceptual meaningfulness and practical usefulness have recently been questioned. In this paper, we take the position that the dichotomy is conceptually meaningful and indispensable for (uncertainty-aware) machine learning. In particular, we argue that much of the recent criticism is flawed, either because it targets specific mathematical frameworks for the dichotomy, or because it is based on implicit assumptions that are not implied by the dichotomy itself. Moreover, we highlight that the dichotomy is indispensable for a broad class of decision-making problems, where optimal performance provably requires a distinction between reducible and irreducible uncertainty. Therefore, while existing methodology should certainly be improved further, there is no reason to question the distinction per se.
Show more
Self-Programmed Execution for Language-Model Agents
Luke J O'Connor
At the heart of existing language model agents is a fixed orchestrator program which is responsible for the state transition between consecutive turns. This paper introduces \emph{self-programmed execution} (SPE), an agent architecture in which the model completion is itself the orchestrator program, and the harness evaluates this program but does not impose its own orchestration policy. I formalize this idea using agentic machines: an SPE state is one from which a model completion can load any state of an embedded copy of the machine, meaning that it is subject to no fixed turn-to-turn orchestration policy. Realizing SPE in practice is nontrivial because the same data is both model context and executable program. I therefore introduce \textsc{Spell}, a Lisp-based language in which programs can edit and re-evaluate themselves, and effectful expressions like model invocations are structured such that re-evaluating an edited program does not replay its side effects. Experiments with existing models, not trained for SPE or \textsc{Spell}, show that frontier models can operate in this regime and accomplish challenging agentic tasks. These results demonstrate how an LM can act as an agent without any fixed orchestration policy, and they raise the question of what self-orchestration strategies might be learned by a model trained for self-programmed execution.
Show more
Releasing Anchors from Cross-View Correspondence: Probabilistic Multi-View Anchor Graph Clustering
Zhoumin Lu ⋅ Yongbo Yu ⋅ Yu Duan ⋅ Feiping Nie ⋅ Yicong Zhou
Anchor graphs are a standard route to scalable multi-view clustering, but most existing pipelines still depend on either a shared anchor set sampled from concatenated features or a separate anchor-alignment stage before fusion. Both choices are restrictive: the former hard-codes equal anchor budgets and importance at anchor-construction time, while the latter introduces a cumbersome correspondence problem and still struggles with unequal anchor numbers. We develop an alignment-free probabilistic formulation in which view-specific anchor graphs are noisy manifestations of a shared latent cluster identity rather than objects that require cross-view correspondence. Each view has its own anchor space, number of anchors and cluster prototypes; the only shared latent variable is the sample label. We instantiate this idea with a Gaussian latent-label model and a deterministic prototype M-step. The resulting variational EM algorithm admits closed-form updates with convergence guarantees, while experiments demonstrate competitive clustering performance and practical runtime on most datasets.
Show more
Colored Noise Diffusion Sampling
Hadar Davidson ⋅ Noam Issachar ⋅ Sagie Benaim
Diffusion models achieve state-of-the-art image synthesis, with their generative trajectories fundamentally exhibit a spectral bias, resolving low-frequency global structures early and high-frequency fine details later. Conventional stochastic differential equation (SDE) solvers fail to account for this dynamic, naively injecting uniform white noise throughout the entire process and misusing the finite energy budget. In this work, we establish a mathematical framework that reconsiders SDE inference as a targeted, frequency-decoupled energy transfer. Leveraging this framework, we introduce Colored Noise Sampling (CNS), a novel, training-free stochastic solver. Rather than injecting uniform white noise, CNS utilizes a dynamic, timestep- and frequency-dependent schedule that more-efficiently allocates injected energy toward structurally unresolved frequency bands. By actively exploiting the model's inherent spectral bias, CNS systematically steers the generated distribution toward the true data manifold. Extensive experiments demonstrate that CNS significantly outperforms standard ODE and SDE baselines as a strictly plug-and-play, inference-time sampler substitution across diverse architectures (SiT, JiT, FLUX). Compared to standard sampling on ImageNet-256, CNS achieves substantial unguided FID reductions, improving from 8.26 to 6.27 on SiT-XL/2, 32.39 to 26.69 on JiT-B/16, and 11.88 to 8.31 on JiT-H/16, while yielding consistent relative FID improvements with Classifier-Free Guidance.
Show more
probly: Uncertainty-Aware Machine Learning
Paul Hofman ⋅ Clemens Damke ⋅ Timo Löhr ⋅ Santo Thies ⋅ Alireza Javanmardi ⋅ Valentin Margraf ⋅ Jakub Paplhám ⋅ Yusuf Sale ⋅ Eyke Hüllermeier ⋅ Maximilian Muschalik
As machine learning systems are increasingly deployed in real-world applications, the question of how to represent and quantify uncertainty has moved from a methodological side issue to a central concern. In practice, however, making a model uncertainty-aware is still surprisingly difficult: relevant tools tend to be scattered across libraries, each tied to a particular framework and a particular approach to modeling uncertainty, and the choice of representation, quantification, and evaluation is typically left to the user without much guidance. In this paper, we present probly, a Python package that addresses theses issues in a single, modular framework. probly offers (i) lightweight transformations that turn existing models into uncertainty-aware ones, currently supporting PyTorch, scikit-learn, and Flax, (ii) several representations, including second-order distributions, credal sets and conformal prediction, (iii) the corresponding quantification measures, and (iv) a number of evaluation protocols, which can be combined more or less arbitrarily. Using probly, we conduct a benchmark study on three standard tasks --- selective prediction, out-of-distribution detection, and active learning. In addition, and unlike existing benchmarks, we evaluate suitable methods on first-order data, i.e., datasets for which the target itself is a distribution over outcomes. We further illustrate the flexibility of the package on a number of less standard case studies, including large language models, graph neural networks, and data streams. The package is publicly available at \url{https://anonymous.4open.science/r/probly}.
Show more
ExComm: Exploration-Stage Communication for Error-Resilient Agentic Test-Time Scaling
Woomin Song ⋅ Beomjun Kim ⋅ Daewon Choi ⋅ Sai Muralidhar Jayanthi ⋅ Saket Dingliwal ⋅ Jinwoo Shin ⋅ Aram Galstyan
A common failure mode in long-horizon agentic test-time scaling is error propagation, where factual errors or invalid deductions introduced at intermediate steps persist in the agent's belief state and contaminate later reasoning. Existing test-time scaling methods provide limited control over this process, as they often rely on agents to detect their own mistakes, select among flawed trajectories, or refine solutions only after errors have already shaped the reasoning path. We propose ExComm, a communication protocol for exploration-stage agentic test-time scaling. ExComm is motivated by the empirical observation that the majority of intermediate errors in parallel agentic reasoning produce detectable cross-agent factual conflicts. Leveraging the iterative structure of agentic workflows, ExComm periodically audits agent belief states to detect such conflicts, resolves them through a dedicated tool-based verification loop, and returns concise, targeted feedback to the involved agents. Corrections are incorporated through soft belief updates, which append verified feedback rather than overwriting existing beliefs. Furthermore, to prevent collapsing trajectory diversity due to communication, ExComm further introduces a trajectory diversification module that redirects redundant trajectories toward orthogonal strategies. Experiments on AIME 2024, AIME 2025, and GAIA with Gemini-2.5-Flash-Lite and Qwen3.5-4B show that ExComm consistently outperforms strong test-time scaling baselines, achieving average performance gains of 5.7\% and 5.0\% over the best-performing baselines, respectively. Further analyses demonstrate improved error recovery, favorable scaling behavior, stronger diversity than adapted communication baselines, and the best performance-cost trade-off among the evaluated methods.
Show more
Fractional State Space Transition for Long Sequence Modeling
Ivan Kobyzev ⋅ Abbas Ghaddar ⋅ Ali Nasiri-Sarvi ⋅ Lifeng Shang ⋅ Yufei Cui
State Space Models (SSMs) compress sequence history into a bounded recurrent state, making the resulting memory law a central architectural choice for long-context performance. Most modern SSMs rely on ODE-based dynamics that lead to exponential forgetting, limiting their ability to retain information over broad temporal ranges. We introduce FRAC, a selective SSM architecture derived from fractional dynamics that replaces this exponential decay with power-law long memory. To make fractional dynamics practical, FRAC approximates the heavy-tailed target kernel with a finite-state, log-spaced sum of exponential modes. This construction turns fractional memory into an efficient recurrent module with parallel training and prefill, while retaining bounded-state autoregressive decoding. Extensive experiments, including 1.3B-parameter language modeling, demonstrate that FRAC consistently improves long-context performance over state-of-the-art SSM baselines while staying competitive on short-context. These results show that fractional dynamics provide a practical and effective prior for long-context SSMs.
Show more
Leaderboard Hacking: Preference-Based Model Evaluations are Vulnerable to Manipulation
Eve Fleisig ⋅ Joachim Baumann ⋅ Dylan Hadfield-Menell ⋅ Dirk Hovy
Arena-style rankings of language models are a widely used evaluation framework in leaderboards like LMArena, research papers, and model evaluation in production. These rankings aim to realistically assess model performance using pairwise comparisons of anonymous models on user prompts, aggregating votes via the Bradley-Terry model to produce the final ranking. However, we find that these leaderboards are highly vulnerable to several manipulation strategies that unscrupulous providers could use to boost a model's rank: strategic voting (individual votes that boost a model's rank, including votes on irrelevant models), strategic prompting (choosing prompts that favor a model), and strategic nomination (boosting the rank of a model by inserting irrelevant models). We demonstrate these effects on public leaderboard data, analyze the circumstances that cause these vulnerabilities, and describe approaches to safeguard against each attack. Notably, these issues hold for any pairwise preference-based evaluation that uses the Bradley-Terry model. These manipulations boost the rank of nearly every tested model; each attack can boost a model by up to 15-21 places in a ranking and, combined, they can boost a model by up to 45 places.
Show more
ParetoSlider: Diffusion Models Post-Training for Continuous Reward Control
Shelly Golan ⋅ Michael Finkelson ⋅ Ariel Bereslavsky ⋅ Yotam Nitzan ⋅ Or Patashnik
Reinforcement Learning (RL) post-training has become the standard for aligning generative models with human preferences, yet most methods rely on a single scalar reward. When multiple criteria matter, the prevailing practice of ``early scalarization'' collapses rewards into a fixed weighted sum. This commits the model to a single trade-off point at training time, providing no inference-time control over inherently conflicting goals -- such as prompt adherence versus source fidelity in image editing. We introduce ParetoSlider, a multi-objective RL (MORL) framework for Diffusion Models post-training that allows continuous reward control during inference. By training the model with continuously varying preference weights as a conditioning signal the model learns how its denoising trajectory should change as the requested reward priority changes. Consequently, we enable users to navigate optimal trade-offs at inference time without retraining or maintaining multiple checkpoints. We evaluate ParetoSlider across three state-of-the-art flow-matching backbones: SD3.5, FluxKontext, and LTX-2. Our single preference-conditioned model matches or exceeds the performance of baselines trained separately for fixed reward trade-offs, while uniquely providing fine-grained control over competing generative goals.
Show more
UECO: A Unified Encoder with Structure-Aware Attention Mixture via Iterative Edge Evolving for Neural Combinatorial Optimization
Wenzheng Pan ⋅ Shuyi Yan ⋅ Nuoyan Chen ⋅ Yiyang Qu ⋅ Jiale Ma ⋅ Junchi Yan
Despite the promise of Neural Combinatorial Optimization (NCO), recent literature has increasingly coupled *decision paradigms*, *training objectives*, and *problem classes* with specific backbone architectures, limiting generic model design and fair cross-paradigm evaluation. We propose UECO, a simple yet effective **U**nified **E**ncoder for **C**ombinatorial **O**ptimization with a structure-aware attention mixture scheme, where a lightweight module defined upon graph-based locality is interleaved with conventional (global) node-focused attention blocks to dynamically evolve edge representations and inject local topological information into attention scores. Instead of fusing static edge features in a single step, UECO thereby integrates local and global messages in a progressive and problem-agnostic manner. UECO is orthogonal to task-specific decoders, and works seamlessly with *Local Construction* (LC), *Global Prediction* (GP), and *Adaptive Expansion* (AE) paradigms, across node- or edge-oriented problems, symmetric or asymmetric tasks, and sparse or dense graphs, for supervised, reinforcement, or generative learning. Extensive experiments show that, via plug-in substitution of encoders, UECO consistently improves solution and generalization quality over diverse backbones and task scales on TSP, ATSP, CVRP, ACVRP, MIS, MaxClique, and MaxCut.
Show more
Spectral Energy Allocation Enables Source-Free Domain Adaptation in Time Series Forecasting
Zhimin Mei ⋅ Gezheng Xu ⋅ Mohammad Hossein Moslemi ⋅ Ruiyi Fang ⋅ Nima Hosseini Dashtbayaz ⋅ Ziyan Wang ⋅ Jincheng Zhou ⋅ Haolin Jiang ⋅ Song Tang ⋅ Charles Ling ⋅ Boyu Wang
Source-free domain adaptation (SFDA) adapts a pre-trained source model to an unlabeled target domain without access to source data. However, existing SFDA studies have predominantly focused on classification tasks, leaving time series forecasting (TSF), a fundamentally different problem with continuous outputs and complex temporal dependencies, largely unexplored. In this work, we identify unique characteristics and key challenges of SFDA in TSF: 1) forecasting labels correspond to future temporal segments of the same underlying signal, enabling us to split target input and construct a self-supervised learning task, thereby training an accurate short-term forecasting model to provide high-quality pseudo-labels; 2) however, such supervision remains inherently *local* rather than *global* for long-horizon source model adaptation. To fill this gap, we propose the notion of Spectral Energy Allocation (SEA) pattern, defined over a short-long signal pair. The SEA pattern provides a structured correspondence that bridges signals across different temporal horizons, enabling reliable short-horizon pseudo-labels to guide long-horizon adaptation. Extensive experiments demonstrate that our method substantially outperforms baselines across different forecasting backbones.
Show more
Inductive Deductive Synthesis: Enabling AI to Generate Formally Verified Systems
Shubham Agarwal ⋅ Alexander Krentsel ⋅ Shu Liu ⋅ Mert Cemri ⋅ Audrey Cheng ⋅ Rui Meng ⋅ Tomas Pfister ⋅ Chun-Liang Li ⋅ Sylvia Ratnasamy ⋅ Aditya Parameswaran ⋅ Matei A Zaharia ⋅ Ion Stoica ⋅ Mohsen Lesani
AI agents increasingly excel at generating, testing, and refining code. However, they fall short on tasks requiring formal guarantees of full coverage that testing alone cannot provide. Distributed systems are a prime example: properties such as consistency between reads and writes must hold under every possible interleaving of events. Mechanized formal verification can guarantee such correctness, but typically demands months to years of expert effort. As evidence, even SOTA coding agents (Claude Code with Opus 4.6, Codex with GPT-5.4) succeed on only 2/7 distributed key-value-store specifications. In this paper, we present the first effective approach to addressing this gap, Inductive Deductive Synthesis (IDS), which jointly and incrementally synthesizes implementation and proof, and learns from failed attempts to systematically try promising strategies. Built as an agentic LLM system, IDS achieves 7/7 in about 6.8 hours and $106 per spec on average, roughly 200× faster than expert effort and 17% cheaper than SOTA agents. IDS further incorporates performance feedback into the same loop, yielding implementations up to 3× faster than published verified systems.
Show more
Mitigating Confidence Miscalibration in Open-World Semi-Supervised Learning
Wenqiang Wu ⋅ Feng Wang ⋅ Jiye Liang ⋅ Liang Bai
Open-world semi-supervised learning aims to use limited labeled data from known classes to classify unlabeled samples that contain both known and novel classes. Most existing methods over-rely on known classes and tend to generalize incorrectly in complex open-world settings. This leads to degraded pseudo-label quality and the model is often overconfident in wrong predictions. To address this, this paper proposes a **M**itigating **C**onfidence **M**iscalibration Method in **Open**-World Semi-Supervised Learning (**OpenMCM**). By leveraging discriminative knowledge from known classes to guide learning in the novel class space, our method effectively mitigates overconfident misclassifications and improves discrimination accuracy. Furthermore, we propose an adaptive threshold calibration strategy to independently determine optimal decision boundary for each class. By integrating a high-reliable pseudo-label fusion mechanism, we enhance recognition stability for novel classes. Experiments on three benchmark datasets ($i.e.$, CIFAR-10, CIFAR-100, and ImageNet-100) show that our framework achieves performance improvements over existing state-of-the-art methods.
Show more
GenAI Evaluation Results are Largely Artifacts of Evaluation Design Choices: Evidence from Audit Studies of Resume Screening
Nicole Meister ⋅ Hannah Cha ⋅ Mohammed Alsobay ⋅ Alexandra Chouldechova ⋅ Alex Dow ⋅ Hanna Wallach ⋅ Jennifer Wortman Vaughan ⋅ Carlos Guestrin ⋅ Tatsunori Hashimoto ⋅ Solon Barocas
Generative AI is designed to support many tasks, many ways of framing those tasks, and seemingly endless variation in task inputs. But the same flexibility that makes these systems powerful also makes them difficult to evaluate reliably. This flexibility exacerbates the numerous, but often underrecognized and consequential, \emph{researcher degrees of freedom} where even seemingly minor experiment design decisions can substantially affect findings. We demonstrate this problem through an analysis of 36 papers on gender and racial bias in generative AI-assisted resume screening. Although these studies address the same broad research question, they reach strikingly divergent conclusions. To explain this divergence, we adapt recent work on integrative experiment design from the social sciences, and decompose the dimensions along which the studies vary. We then run 1,134 counterfactual experiments by combinatorially varying the experimental modules (names, job-resume pairings, and models) across three baseline experimental frameworks. We find the results prove remarkably brittle: a different design choice flips the conclusion of the reproduced experiment around 22\% of the time. For example, seemingly inconsequential design choices, such as the phrasing of a job description, can reverse the observed conclusion. However, this instability is not always purely stochastic. In some settings, we can build a model that predicts the result of unseen counterfactual experiments with a mean held-out predictive $R^2$ of 0.807. This suggests that despite sensitivities to difference in evaluation design, there exists a learnable structure.
Show more
Impossibility of Distribution-Free Predictive Inference for Individual Treatment Effects
Chongguang Tao ⋅ Zheng Zhou ⋅ Yuhong Yang
Uncertainty quantification for individual treatment effects (ITEs) is a daunting challenge in causal inference. Motivated by recent advances in conformal prediction, several works aim to construct distribution-free prediction sets for ITEs with desired coverage under standard assumptions such as strong ignorability and overlap. In this paper, we show that such goals are fundamentally unattainable in the presence of continuous covariates. Specifically, we establish finite-sample and asymptotic impossibility results demonstrating that any distribution-free prediction set achieving desired coverage for ITEs must be trivial, in the sense that it has infinite expected length. Our analysis relies on a connection between ITE inference and the hardness of conditional independence testing, and highlights the intrinsic limitations imposed by the missing data nature of causal inference. These results provide a new perspective on existing methods, clarifying that their apparent success necessarily relies on additional structural assumptions beyond standard causal assumptions.
Show more
Computer Use at the Edge of the Statistical Precipice
Pierluca D Oro ⋅ Sneha Silwal ⋅ William R Wong ⋅ Yuxuan Sun ⋅ Fanyi Xiao ⋅ Manchen Wang ⋅ Eric Gan ⋅ Allen Bolourchi ⋅ Joseph Tighe
Evaluating Computer Use Agents (CUAs) on interactive environments is fraught with methodological pitfalls that the field has yet to systematically address. We show that a 1MB replay script that blindly executes a recorded action sequence without ever observing the screen outperforms frontier models on prominent static benchmarks, and prove that its expected success rate is exactly equal to the source agent's pass@k in deterministic environments. We trace this and other failures to two root causes: non-principled environment design (static, unsandboxed, or unreliably verified environments) and non-principled evaluation methodology (naïve aggregation and misuse of pass@k for stateful UI interactions). To address the first, we propose PRISM, five design principles for CUA environments (privileged verification, realistic environments, integrity-checked configurations, sandboxed execution, and multifactorial variability) and instantiate them in DigiWorld, a benchmark of 15 realistic sandboxed mobile applications able to evaluate agents in over 3.2 million verified unique configurations. To address the second, we develop an aggregation framework pairing Wilson score intervals with hierarchical bootstrap, producing confidence intervals that correctly account for the nested structure of CUA benchmarks, as we empirically demonstrate. All together, we show that principled environment design and rigorous evaluation methodology are not optional refinements but prerequisites for meaningful CUA research.
Show more
Filtered Conformal Ellipsoids for Graph-Native Time Series
Yannick Limmer
Joint prediction sets for multivariate time series should control a single event while adapting to cross-coordinate dependence. We study filtered conformal ellipsoids: a frozen state-space filter emits a one-step predictive mean and covariance, and split-conformal calibration is applied to the resulting Mahalanobis scores. The filter is used to choose the ellipsoid shape; conformal calibration chooses the scalar radius, so the construction benefits from a learned predictive covariance without relying on Gaussian tail probabilities for coverage. The main difficulty is that filtered scores are dependent and learned recurrent filters need not contract in their raw hidden state; we therefore analyse contraction in an observable predictive-law quotient that identifies hidden states producing the same future sequence of emitted Gaussian laws. Under a stable Bayes Gaussian-projection filter, covariance bounds, and a finite-horizon observability/Fisher condition, small excess Gaussian negative log-likelihood implies contraction of the learned emitted laws. Combined with a threshold-autocovariance envelope this yields a Chebyshev-type approximate coverage bound for filtered split-conformal prediction under dependence; a sharper Bernstein-type bound requires an additional geometric-mixing concentration assumption. Under Gaussian oracle realisability we also obtain a near-oracle log-volume comparison within the class of conditionally valid Gaussian ellipsoid rules. We instantiate the framework with a GCN-GRU filter with diagonal-plus-low-rank covariance. On moderate-size graph-native traffic benchmarks (METR-LA-20 and PEMS-BAY-50), the learned filter gives sharper at-target ellipsoids than static-covariance and non-filter baselines; at full-graph scale and on non-graph-native datasets, factor and copula baselines can be stronger.
Show more
Shaping Useful Noise: Energy Distributions Predict Visual Pretraining Quality
Ching Lam Choi ⋅ Antonio Torralba ⋅ Phillip Isola ⋅ Stefanie Jegelka
Synthetic data can scale visual pretraining, but current selection criteria only partially explain which datasets train strong representations. We propose using the energy profile of a dataset---the distribution of log-probability levels it occupies under a normalised real-image density model---as an audit and curation principle for synthetic data. To operationalise this view, we introduce Energy-Level Distribution Distance (Eld), which compares real and synthetic energy distributions. Real images span sparse high-likelihood scenes and dense low-likelihood textures, so useful synthetic data should cover this energy range rather than collapse to likely but low-variation samples. Across diverse synthetic datasets, Eld exposes mismatches missed by FID and predicts pretraining performance beyond FID, LPIPS variation, and feature-space recall, including across CNN and ViT self-supervised settings. We then use Eld to resample and mix synthetic sources, matching the real-image energy histogram while preserving diversity and improving downstream transfer.
Show more
MaxSketch: Robust Distinct Counting in Streams via Random Projections
Nikos Tsikouras ⋅ Constantine Caramanis ⋅ Christos Tzamos
Estimating the number of distinct elements in a data stream is well understood when repeated elements are identical. In modern settings, however, observations are high-dimensional and noisy, so repeated instances of the same object are only approximately similar -- for example, different images of the same individual may vary significantly at the pixel level. Classical sketches such as HyperLogLog rely on consistent hash values for identical elements and break down in this regime. Recent work on robust distinct counting in general metric spaces achieves $\tilde\Theta(\sqrt{n})$ memory, which is tight in the worst case. We show that substantially improved memory guarantees are possible under geometric structure common in learned representations. We introduce MaxSketch, a simple max-linear sketch built from random Gaussian projections, and prove that it succeeds in estimating the number of distinct latent objects. Concretely, we show that under this assumption $m = \tilde{O}(\log n / \varepsilon^4)$ random projections (and hence $\tilde{O}(\log n/\varepsilon^4)$ memory) suffice to recover the true distinct count within a $(1+\varepsilon)$ factor. Experiments on image streams confirm that MaxSketch accurately estimates distinct counts and generalizes beyond the training regime. Our results bridge classical streaming algorithms and modern representation learning, showing how geometric structure can fundamentally reduce the complexity of distinct counting.
Show more
APS: Bias-Controlled Adaptive Prototype Simulation for Population-Scale LLM Agents
Quan ZHENG ⋅ Yan Gao ⋅ shaobin he ⋅ Haoxiang Guan ⋅ Yuanhe Tian ⋅ Jie Feng ⋅ Ming Wang ⋅ Shuxin Zheng ⋅ Zhen Liu
LLM-agent simulation offers a flexible computational tool for studying population response trajectories that depend on scenario events, memory, demographics, and evolving social context. However, full multi-round simulation scales linearly with both population size and horizon, requiring every agent to query the LLM at every round. We propose \textbf{Adaptive Prototype Simulation (APS)}, a framework that treats scaling as a recurrent LLM-oracle allocation problem rather than a static sampling, propagation, or surrogate-modeling problem. APS preserves the specified LLM as the online transition oracle while querying adaptive core prototypes, selected singleton-tail agents, and shadow-audit agents. Prototype responses induce local response surfaces for nearby agents, reducing online LLM calls without replacing the underlying transition model. To control approximation bias, shadow-audit residual correction estimates propagation residuals for aggregate correction and future budget allocation, while tail-protected singleton routing directly queries selected isolated, heterogeneous, or high-curvature regions that are vulnerable to smoothing. We analyze APS as an estimator of the LLM-induced population transition and decompose its error into prototype-coverage error, shadow-audit residual-correction error, local-propagation bias, and temporal context mismatch. Under the reported protocols, APS gives lower reference-aligned distributional discrepancy than scale-oriented and same-budget baselines while reducing online LLM calls, with ablations and compact robustness checks diagnosing the main bias-control mechanisms. In a 10M-agent, multi-round public-opinion simulation, APS achieves a $381.1\times$ reduction over full simulation, with reference-aligned final-round JSD 0.094 against the corresponding full-LLM reference.
Show more
Can LLM Agents Respond to Disasters? Benchmarking Heterogeneous Geospatial Reasoning in Emergency Operations
Junjue Wang ⋅ Weihao Xuan ⋅ Heli Qi ⋅ Pengyu Dai ⋅ Kunyi Liu ⋅ Hongruixuan Chen ⋅ Zhuo Zheng ⋅ Junshi Xia ⋅ Stefano Ermon ⋅ Naoto YOKOYA
Operational disaster response goes beyond damage assessment, requiring responders to integrate multi-sensor signals, reason over road networks, populations and key facilities, plan evacuations, and produce actionable reports. However, prior work largely isolates remote-sensing perception or evaluates generic tool use, leaving the end-to-end workflows of emergency operations underexplored. In this paper, we introduce **D**isaster **O**perational **R**esponse **A**gent benchmark (**DORA**), the first agentic benchmark for end-to-end disaster response: 515 expert-authored tasks across 45 real-world disaster events spanning 10 types, paired with expert-verified, replayable gold trajectories totaling 3,500 tool-call steps. Tasks span five dimensions that cover the operational disaster-response pipeline: disaster perception, spatial relational analysis, rescue and evacuation planning, temporal evolution reasoning, and multi-modal report synthesis. Agents compose calls from a 108-tool MCP library over heterogeneous geospatial data: optical, SAR, and multi-spectral imagery across single-, bi-, and multi-temporal sequences (0.015-10m GSD), complemented by elevation and social vector layers. We comprehensively evaluate 13 frontier LLMs on our benchmark, revealing three persistent challenges: 1) disaster-domain grounding exposes unique failure modes (damage-semantic grounding, sensor-modality mismatch, and disaster-pipeline composition); 2) agents are doubly bottlenecked by tool selection and argument grounding, where gold tool-order hints improve accuracy by only 1.08-4.40%, and alternative scaffolds yield at most a 3.24% gain; 3) compositional fragility scales with trajectory length, the agent-to-gold gap widening from 7% to 56% on long pipelines. DORA establishes a rigorous testbed for operationally reliable disaster-response agents.
Show more
EntiRE: Invariant Learning for Robust Concept Erasure in Text-to-Image Generative Models
Fengyuan Yu ⋅ Yuyuan Li ⋅ Xiaohua Feng ⋅ Li Zhang ⋅ Jiaming Zhang ⋅ Xiang Liu ⋅ Jun Wang ⋅ Chaochao Chen
Large-scale text-to-image diffusion models can inadvertently generate harmful or sensitive content learned from uncurated training data, motivating the development of concept erasure methods that remove undesired concepts from pre-trained models. However, recent studies reveal that erased models remain vulnerable to adversarial prompt attacks that recover the supposedly removed concepts, highlighting the need for robust erasure techniques. In this work, we visualize the text embedding distributions of adversarial prompts and find that the difficulty of erasure is non-uniform. Existing methods effectively erase concepts in easier regions but leave localized residual holes in harder regions, which adversarial prompts consistently exploit. Motivated by this finding, we propose Entire-space Robust Erasure (EntiRE), an end-to-end framework that casts robust concept erasure as an out-of-distribution generalization problem. By adopting an invariant learning formulation regularized by total variation, EntiRE enforces more uniform erasure across the continuous prompt space, suppressing the localized residual regions. Extensive experiments on erasing nudity, artistic style, and object concepts demonstrate that EntiRE consistently outperforms state-of-the-art baselines in both robustness against adversarial prompt attacks and utility preservation.
Show more
Exact Gaussian Moment Matching for Residual Networks: a Second-Order Method
Simon Kuang ⋅ Xinfan Lin
We study the problem of propagating the mean and covariance of a general multivariate Gaussian distribution through a deep (residual) neural network using layer-by-layer moment matching. We close a longstanding gap by deriving exact moment matching for the probit, GeLU, ReLU (as a limit of GeLU), Heaviside (as a limit of probit), and sine activation functions; for both feedforward and generalized residual layers. On random networks, we find orders-of-magnitude improvements in the KL divergence error metric, up to a millionfold, over popular alternatives. On a variational Bayesian neural network, we show that our method attains hundredfold improvements in KL divergence from Monte Carlo ground truth over a state-of-the-art deterministic inference method. We also give an smooth-distance error bound showing that, under regularity assumptions, moment matching removes the leading low-variance errors and propagates higher-order local accuracy through the layers of a network.
Show more
DIAGNO: Diagonal Spherical Neural Operators for Heterogeneous Earth Dynamics Modeling
Herui Li ⋅ Bin Lu ⋅ Haonan Qi ⋅ Lei Zhou ⋅ Luoyi Fu ⋅ Xinbing Wang ⋅ Meng Jin
Accurate forecasting of Earth's dynamic systems is vital in geophysics, where neural operators offer a powerful data-driven paradigm. Standard operators often treat the Earth as a flat Euclidean plane, introducing geometric artifacts. In contrast, Spherical Neural Operators (SNOs) resolve this mismatch by operating within the spherical spectral space, naturally adapting to spherical geometries. However, for computational efficiency, existing SNOs typically rely on the rotation equivariance assumption. This assumption simplifies the transitions of physical states into isolated spectral modal evolutions, thereby over-idealizing the Earth as a homogeneous system. To solve this problem, we no longer rely on the rotation equivariance assumption that isolates spectral modal evolution, but directly derive an explicit modal interaction kernel and propose Diagonal Spherical Neural Operators named DIAGNO. DIAGNO explicitly parameterizes cross-modal interactions through spectral modal decomposition, thereby preserving the inherent heterogeneity of Earth dynamics. Furthermore, leveraging the unique diagonal structure of the spherical spectrum, we design inter- and intra-diagonal interaction mechanisms to natively capture essential zonal and meridional dynamics, respecting intrinsic geophysical regularities. Extensive experiments across simulated fluid dynamics and real-world Earth system reanalysis (including atmosphere and ocean) datasets demonstrate DIAGNO’s superior capability, consistently achieving state-of-the-art performance in multi-step forecasting. Beyond this, we validate DIAGNO’s practical analytical value in professional geophysical domains by assessing its fidelity in capturing climatological anomalies and maintaining physical consistency.
Show more
Free Decompression with Algebraic Spectral Curves
Siavash Ameli ⋅ Chris van der Heide ⋅ Liam Hodgkinson ⋅ Michael Mahoney
Tools from random matrix theory have become central to deep learning theory, using spectral information to provide mechanisms for modeling generalization, robustness, scaling, and failure modes. While often capable of modeling empirical behavior, practical computations are limited by matrix size, often imposing a restriction to models that are too small to be realistic. This motivates the inference of properties of larger models from the behavior of smaller ones. Free decompression (FD) is a recently proposed method for extrapolating spectral information across matrix sizes, but its utility is currently limited by strong assumptions that preclude its implementation on more realistic machine learning (ML) models. We use algebraic spectral curve theory to provide a general FD methodology for spectral densities whose Stieltjes transform satisfies an algebraic relation, a modeling assumption that is more likely to hold in practice. This recasts FD as an evolution along spectral curves which can be readily integrated. Our framework enables the expansion of spectral densities that have multiple or multi-modal bulks, that exist at multiple scales, and that contain atoms, all characteristic of real-world data and popular ML models. We demonstrate the efficacy of our framework on models of interest in modern ML, including Hessian and activation matrices associated with neural networks and large-scale diffusion models.
Show more
Offloading Score: Measuring AI Reliance through Counterfactual Workflows
Vishakh Padmakumar ⋅ Lujain Ibrahim ⋅ Zora Wang ⋅ Jennifer Wang ⋅ Q.Vera Liao ⋅ Diyi Yang
AI tools are increasingly integrated into real-world workflows. However, existing measures of reliance on these tools focus on AI output adoption or on self-reported indicators, rather than how task effort is distributed between users and tools. Here, we introduce offloading score, a measure of reliance that quantifies the fraction of cognitive effort offloaded to an AI tool. Offloading score is simulation-based---we construct a counterfactual workflow by estimating how the user would have completed the task without the tool, and then computing the fraction of steps saved by using the tool. We validate offloading score through intrinsic evaluations of metric validity, and a controlled user study ($n=40$) with developers performing programming tasks using AI tools. We vary time pressure to test whether reliance measures capture the known increase in reliance under time pressure. We show that offloading score detects significantly higher reliance in time-constrained settings ($+43\%$, $p=0.018$), while usage-based and self-reported baseline measures do not distinguish the conditions. We complement this with descriptive insights showing that higher reliance manifests as greater delegation and more direct reuse of AI outputs. Finally, we demonstrate an approach of using offloading score in combination with target outcomes of reliance (e.g., task understanding) to identify when reliance may be (in)appropriate. Our framework offers two contributions: an instrument users can apply to measure and reflect on their own reliance, and a quantitative signal that agent designers can utilize to mitigate overreliance.
Show more
ALTER: An Allen's Algebra-Based Evaluation Framework for Temporal Reasoning of LLMs
Hongkai Zhou ⋅ Yanhui Li
Despite the critical role of temporal reasoning in Large Language Models (LLMs), current synthetic benchmarks fall short: they either introduce information leakage or, upon addressing this issue, merely assess questions with explicit time points, failing to rigorously evaluate implicit temporal reasoning capabilities. To bridge this gap, we propose a formally grounded evaluation framework, ALTER, based on Allen’s Interval Algebra. By combining basic relations into complex networks and converting these structures into natural language forms through an automated pipeline, we construct a validated benchmark of 2,500 question-answer pairs, ensuring a controllable and scalable evaluation environment. We conduct experiments on a representative selection of LLMs. Empirical evaluations on diverse LLMs yield three key findings: (1) identifying performance decline of LLMs in highly complex scenarios involving dense events; (2) isolating complex logical interactions—rather than linguistic comprehension—as an unresolved bottleneck; and (3) revealing a critical evaluation bias induced by narrative chronology, with LLMs exploiting textual shortcuts. In summary, these findings reveal the blind spots of current evaluation paradigms, providing a rigorous foundation for the robust evaluation of complex temporal reasoning.
Show more
Mechanism-Aware Ensemble Conditioning for Data-Limited Emulation of Extreme Events
Isabella Thiel ⋅ Juan M. Bello-Rivas ⋅ Yannis Kevrekidis ⋅ Themis Sapsis
Extreme events in chaotic systems are difficult to learn from short trajectories because they are controlled by transient finite-time instability rather than by frequently observed bulk dynamics. We propose a mechanism-aware conditioning plug-in framework that turns a nudged coarse ensemble into a non-intrusive sensor of local instability geometry. In the small-noise regime, the ensemble covariance aggregates the same finite-time deformation kernels that govern local instability, providing a Jacobian-free proxy for the local amplification structure around a synchronized coarse trajectory. A small FiLM module injects statistics of this ensemble geometry into an otherwise unchanged backbone while leaving the coarse simulator unchanged. We demonstrate this interface in two distinct pipelines: a Transformer-style residual-attention corrector for a controlled low-dimensional chaotic system and a probabilistic recurrent STORN corrector for topographic two-layer quasi-geostrophic (QG) flow. In the low-dimensional benchmark, ensemble covariance directions co-activate with OTD modes and FiLM conditioning improves 99th-percentile exceedance-frequency errors over an identical no-context Transformer baseline. In QG, a fixed ensemble-conditioned FiLM-STORN model trained on only 50 time units substantially improves long-horizon rare-event statistics in the data-limited regime, including density-tail errors, exceedance frequencies, and spatial exceedance-area distributions relative to an unconditioned STORN trained on the same data; on averaged high-threshold exceedance diagnostics, it also outperforms the baseline STORN trained with 20 times more high-resolution data. These results show that local instability geometry is not merely interpretable post hoc, but an actionable conditioning signal for data-efficient rare-event emulation.
Show more
Surprises in Proper Positive-Only Learning
Shai Ben-David ⋅ Farnam Mansouri ⋅ Anay Mehrotra ⋅ Manolis Zampetakis
Binary classification from positive-only samples is a variant of PAC learning in which the learner receives i.i.d. samples from the positive region of an unknown target concept, but is evaluated under the original distribution (which places mass on both positive and negative regions). This model dates back to (Natarajan, 1987, STOC), and the characterization of improper learning is well-known – it even appears in textbooks (Kearns and Vazirani, 1994, Exercise 3.7). The characterization of *proper* positive-only learning, however, has long remained open. In this work, we revisit and settle this question: a concept class is properly learnable from positive-only samples if and only if it has finite VC dimension and satisfies a new combinatorial condition, which we call uniform exterior separability. Together with several separation results, this characterization reveals a surprisingly rich landscape that differs sharply from standard PAC learning: proper and improper learning are separated, randomized and deterministic proper learning are separated, there are classes for which *no* ERM is a learner, and finite VC dimension does not suffice even for non-uniform learning. Along the way, we introduce new combinatorial dimensions that we believe can be of broader interest in learning theory.
Show more
Successful Page Load