Skip to yearly menu bar Skip to main content


Workshop Sat, Dec 6, 2025 • 8:00 AM – 5:00 PM PST

Reliable ML from Unreliable Data

Andrew Ilyas ⋅ Alkis Kalavasis ⋅ Anay Mehrotra ⋅ Manolis Zampetakis
Project Page [ OpenReview] 

Abstract

Video

Chat is not available.

Schedule

Timezone: America/Los_Angeles
9:00 AM
Coffee Break
11:00 AM
Poster Session 1
→ Clean-Label Physical Backdoor Attacks with Data Distillation
Thinh Dao ⋅ Khoa D Doan ⋅ Kok-Seng Wong
→ Approximating Human Preferences Using a Multi-Judge Learned System
Fernando Avalos ⋅ Eitan Sprejer ⋅ Augusto Bernardi ⋅ José de Azevedo Faustino ⋅ Jacob Haimes ⋅ Narmeen Oozeer
→ Testing Noise Assumptions of Learning Algorithms
Surbhi Goel ⋅ Adam Klivans ⋅ Konstantinos Stavropoulos ⋅ Arsen Vasilyan
→ Reliable Active Learning from Unreliable Labels via Neural Collapse Geometry
Atharv Goel ⋅ Sharat Agarwal ⋅ Saket Anand ⋅ Chetan Arora
→ Locks Tested Without Burglars: Using Coding Assistants to Break Prompt Injection Defenses
Atharv Singh Patlan ⋅ Pramod Viswanath ⋅ Prateek Mittal
→ Unlocking Transfer Learning for Open-World Few-Shot Recognition
Byeonggeun Kim ⋅ Jun-Tae Lee ⋅ Kyuhong Shim ⋅ Simyung Chang
→ Positive-Unlabeled Learning for Control Group Construction in Observational Causal Inference
Ilias Tsoumas ⋅ Dimitrios Bormpoudakis ⋅ Vasileios Sitokonstantinou ⋅ Athanasios Askitopoulos ⋅ Andreas Kalogeras ⋅ Charalampos Kontoes ⋅ Ioannis Athanasiadis
→ Do Internal Layers of LLMs Reveal Patterns for Jailbreak Detection?
Sri Durga Sai Sowmya Kadali ⋅ Vagelis Papalexakis
→ From Search to Decision: A Framework for Adversarially Robust Approximate Nearest Neighbor Search
Alexandr Andoni ⋅ Themistoklis Haris ⋅ Esty Kelman ⋅ Krzysztof Onak
→ WASP: A Weight-Space Approach to Detecting Learned Spuriousness
Cristian D Paduraru ⋅ Antonio Barbalau ⋅ Radu Filipescu ⋅ Andrei Nicolicioiu ⋅ Elena Burceanu
→ Evaluating robustness of tabular models under meta-features based shifts
Irina Deeva ⋅ Nargiza Amerkhanova ⋅ Alena Kropacheva
→ Selective Preference Aggregation
Shreyas Kadekodi ⋅ Hayden McTavish ⋅ Berk Ustun
→ Towards Context-Aware Domain Generalization: Understanding the Benefits and Limits of Marginal Transfer Learning
Jens Müller ⋅ Lars Kühmichel ⋅ Martin Rohbeck ⋅ Stefan Radev ⋅ Ullrich Köthe
→ Trust, But Attribute: Tracing Impact of Data on Trustworthiness in Supervised LLM Fine-Tuning
Kumar Shubham ⋅ Nishant Sharma ⋅ Karn Tiwari ⋅ Prathosh AP
→ Check Yourself Before You Wreck Yourself: Selectively Quitting Improves LLM Agent Safety
Vamshi Krishna Bonagiri ⋅ Ponnurangam Kumaraguru ⋅ Khanh Nguyen ⋅ Benjamin Plaut
→ Uncovering Confident Failures: The Complementary Roles of Aleatoric and Epistemic Uncertainty in LLMs
Kimia Hamidieh ⋅ Veronika Thost ⋅ Walter Gerych ⋅ Mikhail Yurochkin ⋅ Marzyeh Ghassemi
→ Diffusion-supplemented Implicit Layers: Operator Smoothing for better Implicit Solvers
Dinislam Gabitov ⋅ Bader Rasheed ⋅ Anastasia Antsiferova ⋅ Dmitriy Vatolin
→ AsFT: Anchoring Safety During LLM Fine-Tuning Within Narrow Safety Basin
Shuo Yang ⋅ Qihui Zhang ⋅ Yuyang Liu ⋅ Yue Huang ⋅ Xiaojun Jia ⋅ Kun-Peng Ning ⋅ Jia-Yu Yao ⋅ jigang wang ⋅ Dai Hailiang ⋅ Yibing Song ⋅ Li Yuan
→ Robust Federated Learning under Heterogeneous Data with Generalized Heavy-Ball Momentum
Riccardo Zaccone ⋅ Sai Praneeth Karimireddy ⋅ Carlo Masone ⋅ Marco Ciccone
→ Reliable Models via Responsiveness Verification
Meredith Stewart ⋅ Seung Hyun Cheon ⋅ Bogdan Kulynych ⋅ Lily Weng ⋅ Berk Ustun
→ Not All Samples Are Equal: Quantifying Instance-level Difficulty in Targeted Data Poisoning
William Xu ⋅ Yiwei Lu ⋅ Yihan Wang ⋅ Matthew Yang ⋅ Zuoqiu Liu ⋅ Gautam Kamath ⋅ Yaoliang Yu
→ Fine-Grained Uncertainty Decomposition in Large Language Models: A Spectral Approach
Nassim Walha ⋅ Sebastian Gruber ⋅ Thomas Decker ⋅ Yinchong Yang ⋅ Alireza Javanmardi ⋅ Eyke Hüllermeier ⋅ Florian Buettner
→ Why is Your Language Model a Poor Implicit Reward Model?
Noam Razin ⋅ Yong Lin ⋅ Jiarui Yao ⋅ Sanjeev Arora
→ When "Competency" in Reasoning Opens the Door to Vulnerability: Jailbreaking LLMs via Novel Ciphers
Divij Handa ⋅ Zehua Zhang ⋅ Seyyedamirhossein Saeidi ⋅ Shrinidhi Kumbhar ⋅ Md Nayem Uddin ⋅ Aswin Ravikumar Rangasamy Veerasamy ⋅ Chitta Baral
→ Inducing Uncertainty for Test-Time Privacy
Muhammad Ashiq ⋅ Peter Triantafillou ⋅ Hung Tseng ⋅ Grigorios Chrysos
→ Towards Trustworthy Amortized Bayesian Model Comparison
Šimon Kucharský ⋅ Aayush Mishra ⋅ Daniel Habermann ⋅ Stefan Radev ⋅ Paul-Christian Bürkner
→ The Silent Judge: Unacknowledged Shortcut Bias in LLM-as-a-Judge
Arash Mari Oriyad ⋅ Mohammad Hossein Rohban ⋅ Mahdieh Soleymani
→ Fairness Implications of GNN-to-MLP Knowledge Distillation
Margaret Capetz ⋅ Yizhou Sun ⋅ Arjun Subramonian
→ Bayesian Decision Making Observing an Expert
Daniel Jarne Ornia ⋅ Joel Dyer ⋅ Nicholas Bishop ⋅ Anisoara Calinescu ⋅ Michael Wooldridge
→ Concept-Based Masking: A Patch-Agnostic Defense Against Adversarial Patch Attacks
Ayushi Mehrotra ⋅ Derek Peng ⋅ Dipkamal Bhusal ⋅ Nidhi Rastogi
→ A Few Bad Neurons: Isolating and Surgically Correcting Sycophancy
Claire O'Brien ⋅ Jessica Seto ⋅ Dipankar Roy ⋅ Aditya Dwivedi ⋅ Sunishchal Dev ⋅ Kevin Zhu ⋅ Sean O'Brien ⋅ Ryan Lagasse
→ On the Interaction of Compressibility and Adversarial Robustness
Melih Barsbey ⋅ Antonio Ribeiro ⋅ Umut Simsekli ⋅ Tolga Birdal
→ Stress-Testing Byzantine Defenses under Data Heterogeneity
Latifa Errami ⋅ Hajar Hammouti ⋅ El Houcine Bergou
→ Few-Shot Knowledge Distillation for Language Models via Counterfactual Explanations
Faisal Hamman ⋅ Pasan Dissanayake ⋅ Yanjun Fu ⋅ Sanghamitra Dutta
→ Cross-Lingual Multimodal Retrieval-Augmented Generation for Open Question Answering in Tamil and Yoruba
Kiran Raja ⋅ Mobareji Abejide ⋅ Arya Ram ⋅ Utkarsh Sharma ⋅ Benjamin Liu ⋅ Kevin Zhu
→ Disarming Strategic Text: Span-Aware Counterfactuals for Robust Content Moderation
Hardik Meisheri ⋅ Zaid Hassan ⋅ Karthik Sankaranarayanan
→ Teaming LLMs to Detect and Mitigate Hallucinations
Demian Till ⋅ John Smeaton ⋅ Peter Haubrick ⋅ Mohammed Saheb ⋅ Florian Graef ⋅ David Berman
→ Adversarially-robust probes for Deep Networks
SIMRAN KETHA ⋅ NUTHAN NAGA SAI MUMMANI ⋅ Niranjan Rajesh ⋅ Venkatakrishnan Ramaswamy
→ Responsible Imputation of User Behavior Surveys via Mask-Aware Transformers
Aman Shukla ⋅ Rishabh Kumar ⋅ Daniel Scantlebury
12:00 PM
Lunch
→ StealthEval: A Probe-Rewrite-Evaluate Workflow for Reliable Benchmarks
Lang Xiong ⋅ Nishant Bhargava ⋅ Jeremy Chang ⋅ Jianhang Hong ⋅ Haihao Liu ⋅ Kevin Zhu
→ Ambient Diffusion Omni
Giannis Daras ⋅ Adrian Rodriguez-Munoz ⋅ Adam Klivans ⋅ Antonio Torralba ⋅ Constantinos Daskalakis
→ Active Slice Discovery in Large Language Models
Minhui Zhang ⋅ Prahar Ijner ⋅ Yoav Wald ⋅ Elliot Creager
→ Robust Fine-Tuning from Non-Robust Pretrained Models: Mitigating Suboptimal Transfer With Epsilon-Scheduling
JONAS NGNAWE ⋅ Maxime Heuillet ⋅ Sabyasachi Sahoo ⋅ Yann Pequignot ⋅ Frederic Precioso ⋅ Christian Gagné
→ RL-Guided Data Selection for Language Model Finetuning
Animesh Jha ⋅ Ananjan Nandi ⋅ Harshit Gupta
→ Quantifying CBRN Risk in Frontier Models
Divyanshu Kumar ⋅ Nitin Aravind Birur ⋅ Tanay Baswa ⋅ Sahil Agarwal ⋅ Prashanth Harshangi
→ Uncertainty as Feature Gaps: Epistemic Uncertainty Quantification of LLMs in Contextual Question-Answering
Yavuz Faruk Bakman ⋅ Zhiqi Huang ⋅ Chenyang Zhu ⋅ Anoop Kumar ⋅ Alfy Samuel ⋅ Daben Liu
→ False Sense of Security: Why Probing-based Malicious Input Detection Fails to Generalize
Cheng Wang ⋅ Zeming Wei ⋅ Qin Liu ⋅ Wenxuan Zhou ⋅ Muhao Chen
→ The Impact of Training Data on Adversarial Robustness
Marco Zimmerli ⋅ Andreas Plesner ⋅ Till Aczel ⋅ Roger Wattenhofer
→ Breaking Bad: Exploring the Dangers of LLM-generated Misinformation from Fringe Social Media
Han Kyul Kim ⋅ Hansea Kim ⋅ Eunjeong Joo ⋅ Andy Skumanich
→ Uncertainty-Aware LLMs Fail to Flag Misleading Contexts
Tianyi Zhou ⋅ Johanne Medina ⋅ Sanjay Chawla
→ Unspoken Hints: Accuracy Without Acknowledgement in LLM Reasoning
Arash Mari Oriyad ⋅ Shaygan Adim ⋅ Nima Alighardashi ⋅ Mahdieh Soleymani ⋅ Mohammad Hossein Rohban
→ Return Augmented Decision Transformer for Off-Dynamics Reinforcement Learning
Ruhan Wang ⋅ Yu Yang ⋅ Zhishuai Liu ⋅ Dongruo Zhou ⋅ Pan Xu
→ Ditch the Denoiser: Emergence of Noise Robustness in Self-Supervised Learning from Data Curriculum
Wenquan Lu ⋅ Jiaqi Zhang ⋅ Hugues Van Assel ⋅ Randall Balestriero
→ From Many Voices to One: A Statistically Principled Aggregation of LLM Judges
Jitian Zhao ⋅ Changho Shin ⋅ Tzu-Heng (Brian) Huang ⋅ Satya Sai Srinath Namburi ⋅ Frederic Sala
→ Domain Generalization: A Tale of Two ERMs
Yilun Zhu ⋅ Naihao Deng ⋅ Naichen Shi ⋅ Aditya Gangrade ⋅ Clay Scott
→ Beyond Text: Multimodal Jailbreaking of Vision-Language and Audio Models through Perceptually-Aware Transformations
Divyanshu Kumar ⋅ Shreyas Jena ⋅ Nitin Aravind Birur ⋅ Tanay Baswa ⋅ Sahil Agarwal ⋅ Prashanth Harshangi
→ Deep Research Brings Deeper Harm
Shuo Chen ⋅ Zonggen Li ⋅ Zhen Han ⋅ Bailan He ⋅ Tong Liu ⋅ Haokun Chen ⋅ Georg Groh ⋅ Philip Torr ⋅ Volker Tresp ⋅ Jindong Gu
→ Forget Less, Retain More: A Lightweight Regularizer for Rehearsal-Based Continual Learning
Lama Alssum ⋅ Hasan Hammoud ⋅ Motasem Alfarra ⋅ Juan Leon Alcazar ⋅ Bernard Ghanem
→ Better Data for Satellite Super Resolution
Miguel Castells ⋅ Salzinger Jules ⋅ Oliver Zendel
→ Obscurable Fishermen
Ekaterina Fedorova ⋅ Chara Podimata ⋅ Constantinos Daskalakis
→ Data Decomposition beyond Splitting for Causal Estimation
Xuelin Yang ⋅ Dhruv Singal ⋅ Rina Friedberg ⋅ Niloy Biswas
→ MPSelectTune: Prompt-type Selection for Fine-tuning improves Concept Unlearning in LLMs
Shubhadip Nag ⋅ Srinjoy Das ⋅ Agniva Saha ⋅ Anushree Ghosh ⋅ Soumi Das ⋅ Tarun Kumar ⋅ Suparna Bhattacharya ⋅ Sourangshu Bhattacharya
→ Don’t Make It Up: Preserving Ignorance Awareness in LLM Fine-Tuning
William Shen ⋅ Xinchi Qiu ⋅ Nicola Cancedda ⋅ Nicholas Lane
→ Ambient Proteins: Training Diffusion Models on Low Quality Structures
Giannis Daras ⋅ Jeffrey Ouyang-Zhang ⋅ Krithika Ravishankar ⋅ William Daspit ⋅ Constantinos Daskalakis ⋅ Qiang Liu ⋅ Adam Klivans ⋅ Daniel Diaz
→ SIVA: Self-Improving Vulnerability Agent
Valentin Walischewski ⋅ Giulio Zizzo ⋅ Kevin Webster
→ Conformal Prediction for Molecular Properties under Label Shift
Hyeonsu Lee ⋅ Juyeon Kim ⋅ Erkhembayar Jadamba ⋅ Seungjin Choi ⋅ Hyunjin Shin
→ Improving Consistency in Retrieval-Augmented Systems with Group Similarity Reward
Faisal Hamman ⋅ Chenyang Zhu ⋅ Anoop Kumar ⋅ Xujun Peng ⋅ Sanghamitra Dutta ⋅ Daben Liu ⋅ Alfy Samuel
→ DPO-PRO: Direct Preference Optimization with Preference Robustness
Cheol Woo Kim ⋅ Shresth Verma ⋅ Mauricio Tec ⋅ Milind Tambe
2:15 PM
Panel
3:00 PM
Coffee Break
→ Strategic Feature Selection
Jivat Neet Kaur ⋅ Divya Shanmugam ⋅ Emma Pierson ⋅ Michael Jordan ⋅ Nika Haghtalab ⋅ Ahmed Alaa ⋅ Serena Wang
→ An Analysis of Causal Effect Estimation using Outcome Invariant Data Augmentation
Uzair Akbar ⋅ Niki Kilbertus ⋅ Hao Shen ⋅ Krikamol Muandet ⋅ Bo Dai
→ Keep It Real: Challenges in Attacking Compression-Based Adversarial Purification
Samuel Räber ⋅ Till Aczel ⋅ Andreas Plesner ⋅ Roger Wattenhofer
→ Automated Generation of Multilingual Jailbreak Prompts
Jonathan Ding ⋅ Will Cai ⋅ Khanak Jain ⋅ Dhruv Nair ⋅ Aditya Naha ⋅ Kevin Zhu ⋅ Vasu Sharma
→ Drawing Reliable Conclusions with Imperfect Synthetic Data
Yewon Byun ⋅ Shantanu Gupta ⋅ Zachary Lipton ⋅ Rachel Childers ⋅ Bryan Wilder
→ Beyond Static Bias: Quantifying Fairness Variability in CheXpert
Ines Ayed ⋅ Gabriel Alcover ⋅ Fernando Alonso-Fernandez ⋅ Antoni Jaume-i-Capó
→ It is Hard to Unlearn Dogged Backdoor Samples in Diffusion Models
An Huang ⋅ Zuobin Xiong ⋅ Muchao Ye ⋅ Junggab Son
→ Is Safety Standard Same for Everyone? User-Specific Safety Evaluation of Large Language Models
Yeonjun In ⋅ Wonjoong Kim ⋅ Kanghoon Yoon ⋅ Sungchul Kim ⋅ Mehrab Tanjim ⋅ Sangwu Park ⋅ Kibum Kim ⋅ Chanyoung Park
→ Cost Efficient Fairness Audit Under Partial Feedback
Nirjhar Das ⋅ mohit sharma ⋅ Praharsh Nanavati ⋅ Kirankumar Shiragur ⋅ Amit Deshpande
→ Probabilistic Framework for Robustness of Counterfactual Explanations Under Data Shifts
Xuan Zhao ⋅ Lena Krieger ⋅ Zhuo Cao ⋅ Arya Bangun ⋅ Hanno Scharr ⋅ Ira Assent
→ FairContrast: Enhancing Fairness through Contrastive learning and Customized Augmenting Methods on Tabular Data
Aida Tayebi ⋅ Ali Khodabandeh Yalabadi ⋅ Mehdi Yazdani-Jahromi ⋅ Ozlem Garibay
→ Reliable Unlearning Harmful Information in LLMs with Metamorphosis Representation Projection
Chengcan Wu ⋅ Zeming Wei ⋅ Huanran Chen ⋅ Yinpeng Dong ⋅ Meng Sun
→ Breaking the Mirror: Activation-Based Mitigation of Self-Preference in LLM Evaluators
Dani Roytburg ⋅ Matthew Nguyen ⋅ Matthew Bozoukov ⋅ Hongyu Fu ⋅ Jou Barzdukas ⋅ Narmeen Oozeer
→ DynamiX: Dynamic Resource eXploration for Personalized Ad-Recommendations
Adam Holeman ⋅ Sohini Roychowdhury ⋅ Mohammad Amin ⋅ Feng Wei ⋅ Bhaskar Mehta ⋅ Sri Reddy
→ On Fairness of Task Arithmetic: The Role of Task Vectors
Laura Gomezjurado ⋅ Hiroki Naganuma ⋅ Kotaro Yoshida ⋅ Takafumi Horie ⋅ Yuji Naraki ⋅ Ryotaro Shimizu
→ BridgePure: Limited Protection Leakage Can Break Black-Box Data Protection
Yihan Wang ⋅ Yiwei Lu ⋅ Xiao-Shan Gao ⋅ Gautam Kamath ⋅ Yaoliang Yu
→ Information-Theoretic Conditions for Chain-of-Thought Monitorability and Methods for Improving It
Usman Anwar ⋅ Tim Bakker ⋅ Cristina Pinneri ⋅ Dana Kianfar ⋅ Christos Louizos
→ FAVAE-Effective Frequency Aware Latent Tokenizer
Tejaswini Medi ⋅ Hsien-Yi Wang ⋅ Arianna Rampini ⋅ Margret Keuper
→ KAIROS: Scalable Model-Agnostic Data Valuation
Jiongli Zhu ⋅ Parjanya Prashant ⋅ Alex Cloninger ⋅ Babak Salimi
→ Failure Prediction Is a Better Performance Proxy for Early-Exit Networks Than Calibration
Piotr Kubaty ⋅ Filip Szatkowski ⋅ Metod Jazbec ⋅ Bartosz Wójcik
→ Robust Multi-task Modeling for Bayesian Optimization via In-Context Learning
Yucen Li ⋅ Samuel Daulton ⋅ Samuel Müller ⋅ Andrew Wilson ⋅ Eytan Bakshy
→ Optimal Fair Learning Robust to Adversarial Distribution Shift
Sushant Agarwal ⋅ Amit Deshpande ⋅ Rajmohan Rajaraman ⋅ Ravi Sundaram
→ Safety by Design: High-Probability Constrained Contextual Bandits
Spyridon Konstantinos Dragazis ⋅ Aldo Pacchiano
→ Not All Splits Are Equal: Rethinking Attribute Generalization Across Unrelated Categories
Firca Nicolae ⋅ Elena Burceanu ⋅ Antonio Barbalau ⋅ Dan Oneata
→ Independence-based Fairness via CvM Regularization
Albert Gimó ⋅ Mariia Vladimirova ⋅ Federico Pavone ⋅ Reda CHHAIBI
→ GUARD: Guiding Unbiased Alignment through Reward Debiasing
Advay Samnerkar ⋅ Sagnik Bhattacharya ⋅ Kailash Ranganathan ⋅ Kevin Zhu ⋅ Ashwinee Panda
→ CroPA++: Exposing Vulnerabilities in Vision Language Models and Enhancing Adversarial Transferability of Cross-Prompt Attacks
Agam Pandey ⋅ Amritanshu Tiwari ⋅ Atharv Mittal ⋅ Sukrit Jindal ⋅ Swadesh Swain
→ Reliable Compositional Editing with Overlap-Aware Attention in Diffusion Models
Salamata Konate ⋅ Hassan Hamidi ⋅ Elham Dolatabadi ⋅ Frank Rudzicz ⋅ Laleh Seyyed-Kalantari
→ Reasoning as an Adaptive Defense for Safety
Taeyoun Kim ⋅ Fahim Tajwar ⋅ Aditi Raghunathan ⋅ Aviral Kumar
→ Minimal Repairs for Learning Over Incomplete Data
Cheng Zhen ⋅ Nischal Aryal ⋅ Arash Termehchy ⋅ Prayoga ⋅ Garrett Biwer
→ ERGO: Entropy-guided Resetting for Generation Optimization in Multi-turn Language Models
Haziq Khalid ⋅ Athikash Jeyaganthan ⋅ Timothy Do ⋅ Yicheng Fu ⋅ Vasu Sharma ⋅ Sean O'Brien ⋅ Kevin Zhu
→ A Guide to Robust Generalization: The Impact of Architecture, Pre-training, and Optimization Strategy
Maxime Heuillet ⋅ Rishika Bhagwatkar ⋅ JONAS NGNAWE ⋅ Yann Pequignot ⋅ Alexandre Larouche ⋅ Christian Gagné ⋅ Irina Rish ⋅ Ola Ahmad ⋅ Audrey Durand
→ Sandbagging in a Simple Survival Bandit Problem
Joel Dyer ⋅ Daniel Jarne Ornia ⋅ Nicholas Bishop ⋅ Anisoara Calinescu ⋅ Michael Wooldridge
→ Rethinking Sparse Autoencoders: Select-and-Project for Fairness and Control from Encoder Features Alone
Antonio Barbalau ⋅ Cristian D Paduraru ⋅ Teodor Poncu ⋅ Alexandru Tifrea ⋅ Elena Burceanu