Skip to yearly menu bar Skip to main content

Workshop Mon, Dec 14, 2026 • 12:00 AM – 9:00 AM AEDT C201

Workshop on Resource-Aware Agentic AI

Kai-Wei Chang ⋅ Jiri Gesi ⋅ Yuxuan Lu ⋅ Weijia Shi ⋅ Dakuo Wang ⋅ Luke Zettlemoyer ⋅ Jiaxin Pei

[ OpenReview]

Abstract

Today’s agents can write and debug code, solve competition-level mathematical problems, and conduct deep research, and are increasingly deployed in everyday products and workflows. However, most of these agents are designed to maximize task performance. They are largely unaware of the resources they consume, including the compute, energy, memory, latency, tool cost, and data required to train them. The Workshop on Resource-Aware Agentic AI brings together academic and industrial experts to explore how to design agents that are both efficient and effective, from training and planning to systems and deployment. Resource awareness is not merely a technical challenge. It is essential to make AI agents accessible to all, not only those with large compute budgets.

Chat is not available.

Schedule

Time zone: Conference local time zone

Agents Retrieve but Do Not Re-Derive: Coherence Costs Concentrate on Conclusion

Suranjan Goswami ⋅ Karan Bharadwaj ⋅ Manu Ghulyani

CRISP: Compressed Reasoning and Inter-Agent Symbolic Protocol for Efficient Multi-Agent LLM Communication

Yun Wang ⋅ Susmit Neogi ⋅ Giridhar NR ⋅ Ramya Toshniwal ⋅ Mohan Sasanapuri

When Does LLM Orchestration Pay Off? A Controlled Evaluation of Accuracy, Resources, and Task Difficulty

Nicolas Leins ⋅ Nico Pelleriti ⋅ Jana Gonnermann-Müller ⋅ Sebastian Pokutta

Adaptive Sequential Retrieval in RAG via Generative Modeling and Reinforcement Learning

Raghuram Bharadwaj Diddigi ⋅ Nikita Kiran ⋅ Swetha Murali ⋅ Dyuthi Vivek ⋅ Manish K Vatsa ⋅ Shubham Verma ⋅ Ravi Naik

Diagnose, Then Select: What Makes Critique-Based Refinement Effective?

Hyun Ryu ⋅ Jihwan Oh ⋅ Sumyeong Ahn ⋅ Se-Young Yun

Blind Spots: Where Small Language Models Fall Behind the Frontier Models on Terminal Tasks

Ashima Suvarna ⋅ Ruta Desai ⋅ Ahmed Elgohary Ghoneim

Breaking the Role: At What Point Does Compressing a Shared KV Cache Degrade Agent Capabilities?

Pranav Sathu ⋅ Rishi Khare

Study or Search? Designing Efficient and Effective Coding Agents for Downstream Tasks

Samad Syed

When Lightweight Routers Fail: Distribution Shift and Deferral in Tool-Calling LLM Agents

Autumn Mapes ⋅ Zhichang Chen ⋅ Jialin Yu ⋅ Marc Boubnovski Martell ⋅ Josefa Stoisser

Windowed-MTP: Removing the Full-Context Draft-KV Tax in Million-Token Agentic Serving

Alagappan Valliappan

Planning over Private State: Local–Remote Delegation for Tool-Calling Agents

Ziwen Li ⋅ Tianshi Li

From Budget Awareness to Budget Control: Confidence-Calibrated Intervention for Agentic AI

Guangtao Zheng ⋅ Aravind Cheruvu ⋅ Mahasweta Chakraborti ⋅ Soumyendu Sarkar

When Parallelism Pays Off: Cohesion-Aware Task Partitioning for Multi-Agent Coding

Xu Yang ⋅ Lunyiu Nie ⋅ Ethan Chandra ⋅ Stanislav Gannutin ⋅ Fangru Lin ⋅ Swarat Chaudhuri

The Physical Cost of Context: Quality-Constrained Resource Accounting for Coding Agents

Zhongming Yu ⋅ Yitong Ma ⋅ Hengjia Yu ⋅ Boqin Yuan ⋅ Xiangye Lin ⋅ Hejia Zhang ⋅ Jishen Zhao

Batching the Reflective Optimization Loop: Parallel Proposals Make GEPA Faster and Better

Jialin Zhang ⋅ Lakshya A Agrawal ⋅ Shangyin Tan ⋅ Donghyun Lee ⋅ Lutfi Eren Erdogan ⋅ Dan Klein ⋅ Koushik Sen ⋅ Alex Dimakis ⋅ Matei A Zaharia

Lightweight Neural ODEs for LLM Safety–Helpfulness Steering

Kevin Chen ⋅ Eric Hanchen Jiang ⋅ Ying Nian Wu

Volume or Coupling? A Scale-Dependent Dissociation in Constraint Recovery of Language-Model Loops

Simin Yuan

Observability Debt in Agent Evaluation: Toward a Telemetry Layer for AgenticOS

Rudra P Singh

AgentServeSim: Serving-System Simulation and Policy Search for LLM Agent Programs

Rakibul Hasan Rajib ⋅ Mengxin Zheng ⋅ Qian Lou

AI-BAAM: AI-Driven Bank Statement Analytics as Alternative Data for Malaysian MSME Credit Scoring

Chun Chet Ng ⋅ Zhen Hao Chu ⋅ Jia Yu Lim ⋅ Low W Zeng

How Do AI Agents Spend Your Money? Analyzing and Predicting Token Cost in Agentic Coding Tasks

Longju Bai ⋅ Zhemin Huang ⋅ Xingyao Wang ⋅ Jiao Sun ⋅ Rada Mihalcea ⋅ Erik Brynjolfsson ⋅ Alex Pentland

On the Clock: Towards Punctual and Productive Time-Budgeted AI Agents

Aaron Wang ⋅ Neelabh Madan ⋅ Uladzislau Sobal ⋅ Matthew Trager ⋅ Michael Kleinman ⋅ Elman Mansimov ⋅ Wei Xia ⋅ Stefano Soatto

Trajectory-Derived Confidence for Reliable, Resource-Aware Clinical Text-to-SQL Agents

Mincheol Song ⋅ Joshua Ward ⋅ Jake Jung ⋅ Guang Cheng

How Much Skill Orchestration Can LLM Agents Automatically Compile Away? Towards Token-Efficient Skill Use

Jiayi Ni ⋅ Zac Wang ⋅ Alex Zhang ⋅ Zheng Liu ⋅ Roopa Prabhu

Reading the Plan: Multi-Horizon Future-Action Prediction from Frozen Hidden States in LLM Agents

Laksh Advani ⋅ Suraj Maharjan ⋅ ISHA CHATURVEDI

Selective and Hybrid-State Handoff for Latent Multi-Agent Systems

Miles Liao ⋅ Aaron Smieszek ⋅ Yuhao Ge

Joules-to-GDP: How Much Economic Value Do AI Systems Produce per Unit of Energy?

Amirreza Zeinali ⋅ Jon Saad-Falcon ⋅ Avanika Narayan ⋅ Ramya Ramakrishnan ⋅ Amanda Dsouza ⋅ Paroma Varma ⋅ Vincent Chen ⋅ Azalia Mirhoseini ⋅ Christopher Ré

Agent-Conditioned Routing: Diverse Reasoning Paths Within One Frozen Model

Ananth Kalyanasundaram ⋅ Ankush Kadu ⋅ Aswanth Krishnan

Failure Signals Are Control State: Resource-Aware Retry Scheduling for Lean Agents

Arya S Reddy

MARGENT: Measuring the Marginal Value of Delegation in Agentic Systems

Yuning Yang ⋅ Jingxuan Li ⋅ Yizhou Zhao ⋅ Linfan Zhang ⋅ Longhao Jin ⋅ Manling Li ⋅ Han Liu ⋅ Ying Nian Wu

Spending the Dynamics Budget with Causally Factorized World Models

Lipeng Zu ⋅ Shangqian Gao ⋅ Xiaonan Zhang

Belief Context Graph: A Revisable State for Long-Horizon Agents

Xiaobo Wang ⋅ Shuyi Zhang ⋅ Xinyi Wen ⋅ Keshu Fu ⋅ Jiaqi Li ⋅ Jun Bai ⋅ Yang Liu ⋅ Zixia Jia ⋅ Zilong Zheng

Recursion as Window Relief: A Controlled 2×2 Factorial of Recursive Language Models and Prompt Compression

Evandro Franco ⋅ Akarsha Sehwag

Prefill, Decode, and Context: A Measured Energy Law for Agentic LLM Serving

Vince Wu ⋅ Jiaqi Han ⋅ Boris Ruf ⋅ Vincent Grari ⋅ Marcin Detyniecki

MiniCorp: The Last Mile of the AI Agent Firm

Jingying Zeng ⋅ Zhenwei Dai ⋅ Jinjin Tian ⋅ Jinning Li ⋅ Qi He

VAM: Verbalized Action Masking for Controllable Exploration with Verifiable Rewards

Zhicheng Zhang ⋅ Ziyan Wang ⋅ Yali Du ⋅ Fei Fang

Agentic ESOpt: Fine-Tuning Long-Horizon LLM Agents with Minimal GPU Requirements

Zhi Zheng ⋅ Rongsheng Chen ⋅ Yunpeng Ba ⋅ Zhenkun Wang ⋅ Yee Whye Teh ⋅ Wee Sun Lee

Cognitive State Compilation: Resource-Bounded Construction of Reasoning States for Long-Horizon LLM Agents

Keshav Jain ⋅ Vansh Singh ⋅ Rishi Ashish Shah ⋅ Shivaay Dhondiyal

Metered or Provisioned? Joint Routing and Capacity Planning for Resource-Aware Inference

Zongshun Zhang ⋅ Sara Soltaninejad ⋅ JingHongYu Bian ⋅ Yuan Y Gu ⋅ Grzegorz Kakareko

PRICE: Adaptive LLM Test-time Compute under a Token Budget and Its Closed-form Pareto Frontier

Zhenyu Wang ⋅ zhu ⋅ Yifan Hu

My Budget Is Yours: Learning When to Hand Off in LLM Planner-Executor Teams

Xiang Li ⋅ Wentao Zhang ⋅ Shariyar Murtaza ⋅ Junaid A Bhatti ⋅ Yifan Nie ⋅ Eugene Wen ⋅ Jian Zhao ⋅ Kate Larson

Which Model for Which Role? Cost-Aware Pareto Search over LLM Agent Configurations

Qian Xie ⋅ Wenyue Hua ⋅ Sripad Karne ⋅ Yueli He ⋅ Armaan Agrawal ⋅ Nikos Pagonas ⋅ Eugene Wu ⋅ Kostis Kaffes ⋅ Nairen Cao ⋅ Tianyi Peng

Not All Turns Are Equally Hard: Adaptive Thinking Budgets For Efficient Multi-Turn Reasoning in Agents

Neharika Jali ⋅ Anupam Nayak ⋅ Gauri Joshi

Where Edge VLA Latency Actually Goes: Measured Scaling Laws and Staleness Admissibility for a 450M Vision–Language–Action Model

Prarabdh Misra ⋅ Mohd Ariful Haque

Beyond Tool-Call Count: Cost-Aware Tool-Use Routing for Black-Box LLM Agents

Zhichang Chen ⋅ Autumn Mapes ⋅ Jialin Yu ⋅ Josefa Stoisser ⋅ Marc Boubnovski Martell

Scaling Offline Inference Compute Improves Online Inference Efficiency

Shayan Talaei ⋅ Jirat Chiaranaipanich ⋅ Amirreza Zeinali ⋅ Azalia Mirhoseini ⋅ Amin Saberi

AgentKVShift: Efficient KV Cache Reuse for Agentic Memory Systems

Nilesh Pandey ⋅ Jason Kong ⋅ Lanxiang Hu ⋅ Quanling Zhao ⋅ Yujie Zhao ⋅ Onat Gungor ⋅ Hao Zhang ⋅ Tajana S Rosing

Learning Latency-Aware Orchestration for Multi-Agent Systems

Xi Shi ⋅ Mengxin Zheng ⋅ Qian Lou

Which Features Matter? Memory-Efficient Compression for Cross-Model KV Mappings

Rishi Khare ⋅ Kyle Kam ⋅ Jiayi Qian ⋅ Zishen Wan

Where Does the Energy Go? Profiling LLM Agent Inference on Blackwell GPUs

Qi Luo ⋅ Kunlin Li ⋅ Ziwen Wang ⋅ Yun CHEN