Accepted Papers

1. Talk 1

   Controlling False Discoveries via Auditing Scientific Agents

   Ziang Liu, Ryan Ye, Yuan Sui, Lauren Hyoseo Yoon, Adriane Otopalik, Yisong Yue, Kristin Branson, Jennifer J. Sun

2. Talk 2

   NatureBench: Can Coding Agents Match the Published SOTA of Nature-Family Papers?

   Yuru Wang, Lejun Cheng, Yuxin Zuo, Sihang Zeng, Bingxiang He, Che Jiang, Junlin Yang, Yuchong Wang, Kaikai Zhao, Weifeng Huang, Kai Tian, Zhenzhao Yuan, Yu Fu, Jincheng Zhong, Weizhi Wang, Ning Ding, Bowen Zhou, Kaiyan Zhang

3. Talk 3

   When AI Agents Fall Short: Evaluating Computer-Use Agents for Nonvisual Interaction

   Satwik Ram Kodandaram, Monalika Padma Reddy, Xiaojun Bi, Jiawei Zhou, I. Ramakrishnan, Vikas Ashok

4. Talk 4

   ProgramBench: Can Language Models Rebuild Programs From Scratch?

   John Yang, Kilian Lieret, Jeffrey Jian Ma, Parth Thakkar, Dmitrii Pedchenko, Sten Sootla, Emily McMilin, Pengcheng Yin, Rui Hou, Gabriel Synnaeve, Diyi Yang, Ofir Press

5. Talk 5

   CausalGame: Benchmarking Causal Thinking of LLM Agents in Games

   Zhenhao Chen, Yongqiang Chen, Chenxi Liu, Junchi Yu, Xiangchen Song, Zijian Li, Jialin Li, Philip Torr, Bo Han, Kun Zhang

6. Talk 6

   E-valuator: Reliable Agent Verifiers with Sequential Hypothesis Testing

   Shuvom Sadhuka, Drew Prinster, Clara Fannjiang, Gabriele Scalia, Bonnie Berger, Aviv Regev, Hanchen Wang

7. Talk 7

   Emerging Trends in AI Agent Evaluation

   Felix Chen, Stephan Rabanser, Peter Kirgis, Nitya Nadgir, Rishi Bommasani, Magda Dubois, Elham Tabassi, Cozmin Ududec, Anka Reuel, Sanmi Koyejo, Sayash Kapoor, Arvind Narayanan

8. Talk 8

   Rethinking Psychometric Evaluation of LLMs: When and Why Self-Reports Predict Behavior

   Rafal Kocielnik, Pengrui Han, Peiyang Song, Myrl G Marmarelis, Ramit Debnath, Dean Mobbs, Anima Anandkumar, R. Michael Alvarez

9. Talk 9

   What Benchmark Scores Alone Can't Tell You About Self-Evolving RL

   Haonan Wang, Yusen Zhang, Jiaxiang Liu, Eugene Wu

10. Talk 10

    Who Checks the Citations? Benchmarking Legal Hallucination Detection

    Patty Liu, Dominik Stammbach, Peter Henderson

11. Talk 11

    Fast but False Progress on Benchmarks with Richer Feedback

    Youssef Allouah, John Duchi, Sanmi Koyejo

12. Talk 12

    LLMs Corrupt Your Documents When You Delegate

    Philippe Laban, Tobias Schnabel, Jennifer Neville

Poster session 1 (32)

1. Poster 1

   Low-Cost Assays for Measuring Model Behavior Across Vendors and Releases

   Tapan Parikh

2. Poster 2

   Selection Is Not Evaluation: Independent Continuations for Runtime Intervention in Language Agents

   Mark Kashirskii, Aleksei Stepin, Mikhail Mozikov, Vladislav Kulikov, Boris Tolstokulakov, Mirzoeva Elena I., Ilya Makarov

3. Poster 3

   Canaries for Agent Graders: Which Check Enforces Which Contract, in One Pipeline and Four Public Lean Checkers

   Rob Sneiderman

4. Poster 4

   DeployBench : Benchmarking LLM Agents for Research Artifact Deployment

   Yuanli Wang, Yaoyao Qian, Yue Zhang, Hanhan Zhou, Jindan Huang, Tianfu Fu, Qiuyang Mang, Huanzhi Mao, Wenhao Chai, Wendong Fan, Liqiang Jing

5. Poster 5

   AlphaDiana: A Measurement System for Agents in Verifiable Reasoning Tasks

   Zhanke Zhou, Zongze Li, Weikai Huang, Xuan Li, Chentao Cao, Xiao Feng, Xiangyu Lu, Jinbo Hu, Yi Xie, Nico Pelleriti, Shiyang Liu, Menghan Lu, Max Zimmer, Brando Miranda, Jiangchao Yao, Bo Liu, Sanmi Koyejo, Sebastian Pokutta, Bo Han

6. Poster 6

   Can AI Agents Synthesize Scientific Conclusions in Health?

   Hayoung Jung, Pedro Viana Diniz, José Reinaldo Corrêa Roveda, Abner Fernandes da Silva, Haeun Jung, Enoch Tsai, Aleksandra Korolova, Manoel Horta Ribeiro

7. Poster 7

   DEBATE: A Large-Scale Benchmark for Evaluating Opinion Dynamics in Role-Playing LLM Agents

   Yun-Shiuan Chuang, Ruixuan Tu, Chengtao Dai, You Li, Smit Vasani, Binwei Yao, Michael Henry Tessler, Sijia Yang, Dhavan V. Shah, Robert D. Hawkins, Junjie Hu, Timothy T. Rogers

8. Poster 8

   Auditing the Winner: An Upper Bound on What Critic Disagreement Adds Beyond the Selector on SWE-Agent Trajectories

   Joseph M. Morante Jr.

9. Poster 9

   Which Model for Which Role? Cost-Efficient Pareto Recommendation for LLM Agentic Workflows

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

10. Poster 10

    EnterpriseBench: Evaluating LLM Agents on End-to-End Spreadsheet Tasks in Finance

    Thomson Yen, Julian Poeltl, Harshith Srinivas Gear, Yilin Meng, Joshua Fan, Adam Shen, Yili Liu, Ali Bauyrzhan, Siri Du, Haoyang Liu, C. Daniel Guetta, Hongseok Namkoong

11. Poster 11

    When the Conclusion Breaks: Evaluating Claim Revision in Scientific Agents

    Sarvesh Tiku, Gauri Sharma

12. Poster 12

    Self-Direction: Measuring and Correcting Agent Mistakes Across Sessions

    Yuval Kansal, Karen Ullrich, Polina Kirichenko, Ilia Kulikov, Niraj Jha, Mohammad Pezeshki, Mark Ibrahim

13. Poster 13

    Qualifying AI Agents Under Cost-Aware Human Oversight

    Veronica Chatrath, Bryan Zhu, Jingxuan Fan, George Pu, Soham Dinesh Tiwari, Soham Dan, Ryan Young, Varun Ursekar, Christy Yuan Li, Yuang Yao, Apaar Shanker, Vijay S Kalmath, Minglai Yang, Daniel Yue Zhang, Yunzhong He, Ying Liu, Chenguang Wang, Zhijun Yin, Yuan Xue

14. Poster 14

    Babysitting AI Agents: A User-Centered Taxonomy and Study of Agentic Delegation Failures

    Shravika Mittal, Sunnie S. Y. Kim, Scott Counts, Jennifer Neville, Debarati Das, Philippe Laban

15. Poster 15

    MBABench-2.0: Evaluating AI Agents on Constructing Financial Spreadsheets Beyond Simple Verifiability

    Thomson Yen, Julian Poeltl, Patrick Shea, Shubham Anil Varma, Dhruv Rastogi, Mengshi Zhang, Zihao Chen, Hongseok Namkoong

16. Poster 16

    Beyond Final Answers: Trace-Level Evaluation of Multi-Agent Robustness

    Anna Mazhar, Muhammad Huzaifa Khan Suri, sainyam galhotra

17. Poster 17

    JANUS: Evaluating Goal-Conditioned Distortion in Factually Grounded LLM Communication

    Polydoros Giannouris, Mohsinul Kabir, Sophia Ananiadou

18. Poster 18

    agent-vitals: White-box Evaluation and Monitoring of Agent Harnesses Under Evolving Context

    Ye Ziwei

19. Poster 19

    Synthetic Stakeholders? Agreement and Stability of Open-Weight LLM Agent Panels in Participatory Model Selection

    Tereza Blazkova

20. Poster 20

    Can Small Databases Preserve SQL Evaluation Signals?

    Kate Gwimm, Carson Eisenach

21. Poster 21

    Noise-Calibrated Conclusion Robustness: Separating Manipulation from Variance in Financial Research Agents

    Zixin Audrey Cai

22. Poster 22

    Beyond Isolated Biases: How LLM-Judge Surface Sensitivities Compose

    Jiya Sheetal Rathi

23. Poster 23

    Did the Role Fire? Auditing the Manipulation Checks Behind Role-Prompted Multi-Agent Systems

    Kevin Zhou, Sparsh Roy, Maxim Kudryashov

24. Poster 24

    The Hard Part Comes After Search: Benchmarking Web Agents on Synthesizing, Organizing, and Displaying Knowledge

    Alexander Gill, Md Farhan Ishmam, Xuyen Nguyen, Neha Bhat, Parker Henry DeYoung, Fateme Hashemi Chaleshtori, Nathan Stringham, Kenneth Marino, Ana Marasovic

25. Poster 25

    Evaluating autoresearch agents without verifiable rewards: a case study

    Xinhui Li, Zijiao Chen

26. Poster 26

    Agents Catching Agents: Evaluating Shortcut Cascades and Silent Benchmark Gaming in Clinical Multiagent Committees

    Sebastián Andrés Cajas Ordóñez, Yehudhah Kennedy Rodríguez Moran, Agastya Munnangi, Aldo Marzullo, Felipe Ocampo Osorio, Quang Bui, Mohammad Shahin, Armaan Grewal, Emmanuel Paul Kwesiga, Anqi Peter Li, Josephine Nanyonjo, Aaditya Panchal, Arshnoor Bhutani, Nikhil Jaiswal, Milit S. Patel, Maximin Lange, Renato Umeton, Leo Anthony Celi

27. Poster 27

    Iterated creation, grading, and revision of data science projects by language models

    Rohan Alexander

28. Poster 28

    Decision-Model Controllers Do Not Make Terminal Agents Cheaper: A Placebo-Controlled Replication

    Naman Vats

29. Poster 29

    Agent Plasticity: Measuring Self-Improvement Through Experience

    Harman Singh, Anton Bakhtin, Rulin Shao, Gabriel Synnaeve, Ilia Kulikov, Rob Fergus, Sanjeev Arora, Kurt Keutzer, Anuj Mahajan, Anirudh Goyal

30. Poster 30

    Which Citation Grader? The Measured Hallucination Rate of Research Agents Depends on How References Are Checked

    Faris Raza

31. Poster 31

    World of SWECraft: Agent Collaboration at Scale via Simulated Software Worlds

    Shannon Zejiang Shen, Yapei Chang, Qiyao Liang, Benjamin Newman, Jenny Liang, Rulin Shao, Ziqiao Ma, Junjie Oscar Yin, Yung-Sung Chuang, Weijia Shi, Yejin Choi, Luke Zettlemoyer, Pang Wei Koh, Mohit Iyyer, Tongshuang Wu, David Sontag

32. Poster 32

    Can AI Agents Improve Models Without Making Them Unsafe? Toward Safe Recursive Self-Improvement

    Collin Zhang, Tingwei Zhang, Vitaly Shmatikov

Poster session 2 (32)

1. Poster 1

   Evaluate the Deployed Path: Continual Evaluation Finds Coverage Drift in an Enterprise Agent Governance Layer

   Shaishav Pindadi

2. Poster 2

   The Next Challenge for Agentic Cybersecurity: A Realistic, Contamination-Free Reverse Engineering Benchmark

   Jeremy Spence, Nicholas Assaderaghi, Feng Xiao, Jinhao Zhu, Nikil Ravi, Xiangyu Qi, Matthew Jagielski, Raluca Ada Popa, Eric Wallace, Guannan Wei, Yangruibo Ding, Zhuo Zhang

3. Poster 3

   Time to REFLECT: Can We Trust LLM Judges for Evidence-based Research Agents?

   Leyao Wang, Yanan He, Peng Chen, Asaf Yehudai, Yixin Liu, Rex Ying, Michal Shmueli-Scheuer, Arman Cohan

4. Poster 4

   Can Coding-Agent Leaderboards Tell Whether a Harness Advantage Is Model-Specific? An Audit of Terminal-Bench and SWE-bench

   Yuanyuan Tan, Yingxin Jia, Youwen Zhu, Weijia Han, Shijin Wang, Jinming He

5. Poster 5

   Paying Attention: The Challenges of Expert-Grounded Agent Evaluation

   Andre Shiu, Nathaniel Anleitner, Igor Molybog, Daniel Port, Narayana Santhanam

6. Poster 6

   Better, Faster, Stronger: Programmatic Skill Learning Best Reduces Agent Cost

   Zixi Huang, Xiheng Wang, Andrew Wang, William Jurayj, Bernal Jiménez Gutiérrez, Daniel Khashabi, Nicholas Andrews

7. Poster 7

   Human-Like Generalization of Risk Preferences in LLMs

   Sara Fish, Felix Anthony Sosa, Sonia Krishna Murthy

8. Poster 8

   Accurate but Not Humble: Evaluating Epistemic Humility in LLM Agents under Knowledge Conflict

   Kaiser Sun, Bernal Jiménez Gutiérrez, Hongjun Liu, Jingyu Zhang, Jie Gao, Mark Dredze, Daniel Khashabi

9. Poster 9

   OpenEstimate: Evaluating LLM Priors for Probabilistic Estimation on Real-World Data

   Alana Marzoev, Jillian Ross, Mike Cafarella, Jacob Andreas

10. Poster 10

    Threat-Preserving Representation Sensitivity in Agent-Security Benchmarks

    Neeraj Karamchandani, Piyush Nagasubramaniam, Xinhong Xie, SENCUN ZHU, Dinghao Wu

11. Poster 11

    Harness-Delta Attribution: Dissecting Gains From Harness Evolution

    Wenxuan Ding, Qin Lu, Changlong Yu, Sha Li, Shuowei Jin, Xin Liu, Zhenyu Shi, Greg Durrett

12. Poster 12

    SWE-sweep: Can Agents Autonomously Find and Fix Bugs?

    Kilian Lieret, Jeffrey Jian Ma, Rahul Kindi, Yuxiang Wei, Jeremy Ma, Sten Sootla, Parth Thakkar, Chao Beyond Zhou, Pengcheng Yin, Rui Hou, Ofir Press, John Yang

13. Poster 13

    Accounting for Every Agent-Second in LLM Teams

    Xiang Fu

14. Poster 14

    Persuaded, Not Informed: How Trajectories Bias LLM Judges of Coding Agents

    Kiera McCormick

15. Poster 15

    How State-Transition Fidelity Affects Financial Agent Evaluation

    Gauri Sharma, Sarvesh Tiku

16. Poster 16

    OpenApps: A Simulation Engine for Reliable Computer-Use Agents

    Karen Ullrich, Jingtong Su, Arjun Subramonian, Claudia Shi, Amir Bar, Ivan Evtimov, Nikolaos Tsilivis, Julia Kempe, Randall Balestriero, Mark Ibrahim

17. Poster 17

    What Does a MemoryAgentBench Score Measure? Auditing the Harness and the Published Claims

    Yitao Sun

18. Poster 18

    Emergence World: Adversarial Stress-Testing of Long-Horizon Multi-Agent Systems

    Deepak Akkil, Karthik Vikram Kuduva Gopal Kabirdoss, Tamer Abuelsaad, Matthew Pace, Aditya Vempaty, Saahir Beotra, Ravi Kokku, Satya V. Nitta

19. Poster 19

    Evaluating a Terminal Task Is Itself a Terminal Task: What LLM Graders Need to Judge Agent Work

    Tiancheng Ge, Jiakai Xu, Zhou Yu, Eugene Wu, Kostis Kaffes

20. Poster 20

    Who Took the Hint? Applying Methods of Causal Inference to Evaluate Chain of Thought Monitorability

    Alexis J Diamond

21. Poster 21

    Accurate Diagnosis and Safe Mitigation: A Benchmark for DevOps Agents

    Zhehui Huang, Shlomo Tannor, Suhaib Abdurahman, Amir Abbas Tahmasbi, Patrick Blöbaum, Purak Jain, Xing Niu

22. Poster 22

    Same Data, Different Answers: Shortcuts and Sensitivity in Coding Agents

    Joyce Fang

23. Poster 23

    Exploring Mismatches Between Scientific and Operative Causal Models in LLM Drug-Induced Arrhythmia Reasoning

    Haomiao Chen, Suho Lee, Muhammad Ubadah Tanveer, Timothy Zhou, Manjari Narayan

24. Poster 24

    Auditing Acceptance Decisions in Self-Improving Agent Loops

    Shaurya Beriwala

25. Poster 25

    Representational Control over Self-Report & Behavior Coherence in LLM Risk-Taking

    Rafal Kocielnik, Peiyang Song, Pengrui Han, Myrl G Marmarelis, Ramit Debnath, Dean Mobbs, R. Michael Alvarez

26. Poster 26

    Known-groups validity is not enough for evaluation-awareness detectors

    Ariel Arevalo-Alvarado, Antonio Badilla-Olivas

27. Poster 27

    GDP.xlsx: Benchmarking Language Agents on Professional Workbooks

    Sushant Mehta

28. Poster 28

    ForkSCOPE: Charting the Agentic Garden of Forking Paths

    Tian Zheng, Batuhan Duru Yeltekin, Arjun Balaji

29. Poster 29

    SciBench:Review - Toward Continuous Evaluation of AI Scientific Reviewers

    Tom Cohen, Jiaxin Pei, Julia Leviev

30. Poster 30

    DeepFact: Co-Evolving Benchmarks and Agents for Deep Research Factuality

    Yukun Huang, Leonardo F. R. Ribeiro, Momchil Hardalov, Bhuwan Dhingra, Markus Dreyer, Venkatesh Saligrama

31. Poster 31

    Communication Failure in Distributed Runtime Agent Coordination

    Xisen Wang, Yaqi Zhao, Zijian Liu, Yu Chen, Yi Li, Yusen Zhang, Eugene Wu, Jindong Gu

32. Poster 32

    Colosseum: Auditing Collusion in Cooperative Multi-Agent Systems

    Mason Nakamura, Abhinav Kumar, Saswat Das, Sahar Abdelnabi, Saaduddin Mahmud, Nando Fioretto, Shlomo Zilberstein, Eugene Bagdasarian