1. #0002
   Compliance2LoRA: On-Demand Safety Alignment on Arbitrary Policy Subsets via Hypernetwork-Generated LoRA Adapters Pankayaraj Pathmanathan, Furong Huang
   OpenReview
2. #0004
   Correct Scores, Wrong Conclusions: Auditing Forecasting Benchmarks on Chaotic Dynamics Sriman Narayan Kandi, Trisanth Srinivasan
   OpenReview
3. #0007
   What Carries the Calibration Signal in Incremental Question Answering? A Pre-Specified Confirmatory Decomposition Sarthak Tayal
   OpenReview
4. #0011
   Ranking Under Biased Missingness: When Model Rank Is Not Identifiable from Sparse Leaderboards Jung Min Kang
   OpenReview
5. #0014
   MDI: The Minimum Detectable Improvement of an LLM Judge Hyeong-seob Kim
   OpenReview
6. #0018
   Reinforcement Learning on Benign Facts Amplifies Leakage of Memorized Private Data Renfei Zhang, Niloofar Mireshghallah
   OpenReview
7. #0019
   PopResume: Causal Fairness Evaluation of LLM/VLM Resume Screeners with Population-Representative Dataset Sumin Yu, Juhyeon Park, Taesup Moon
   OpenReview
8. #0020
   The Calibration Channel Determines the Bayes-Error Proxy: An Exact Law for Temperature-Induced Distortion Shreyas Pradeepkumar Khandale
   OpenReview
9. #0021
   Can We Trust Gating Weights? Auditing the Validity of Explanations in Latent-Dynamics Mixtures Takayuki Shinohara
   OpenReview
10. #0024
    Can Model Similarity Verify a Machine Unlearning Action? Evidence from Cross-Auditor and Retraining Controls Anton Sokolov
    OpenReview
11. #0027
    Position: AI Evaluations Should be Grounded on a Theory of Capability Nathanael Jo, Ashia C. Wilson
    OpenReview
12. #0030
    Beyond TSTR: Tabular Foundation Models as Zero-Shot Evaluators of Synthetic Data Quality Yan Li, Anders Krogh
    OpenReview
13. #0032
    Correct Labels Are Not Enough: Reliable Metadata Can Silently Control Model Behavior Renee Jia, Jinzhou Wu
    OpenReview
14. #0038
    Weakly Discriminative, Not Repaired by Profile Injection, and Manipulable: Self-Confidence in LLM Evaluation Vadym Chernets
    OpenReview
15. #0041
    Metrics That Write Themselves: Evolving an Evaluator from Its Own Blind Spots Xing Zhang, Yanwei CUI, Guanghui Wang, Zhihao Lin, Peiyang He
    OpenReview
16. #0045
    Equal Questions, Unequal Allocation: When Cross-Language Cascade Metrics Change Meaning Ranjita Naik
    OpenReview
17. #0048
    Similar Scores, Different High-Risk Episodes: Auditing Model Selection in Longitudinal EHRs Polina Korobeinikova, Dmitry Lvov, Ilya Pershin
    OpenReview
18. #0051
    Useful or Merely Different? Paired-Control Evaluation for Neural Conditioning of Language Models Angelina Parfenova
    OpenReview
19. #0054
    Are We Measuring Anticipation?Auditing Privileged Information in Procedural Video Evaluation Mahsa Mohammadi Eshkaftaki, Sareh Rowlands
    OpenReview
20. #0056
    What Breaks When Facts Become Stories? The Distortions Source-Agreement Metrics Miss When Models Retell Real Events KWANGSUK PARK, JiWoong Yang, Vaniushkina Kseniia
    OpenReview
21. #0057
    Q/R/M/C: An Intervention-Validated Evaluation Protocol for Memory-Based Agents Zamotaev Rodion Sergeevich, Natalia Gusarova, Ivan Tomilov, Aleksandra Vatian
    OpenReview
22. #0060
    Does Benchmark Correlation Predict Selection Quality for Protein Design? Saanvi Skanda Subramanian
    OpenReview
23. #0062
    When the Judge Drifts: Separating Safety Behavior from Its Measurement Along a GRPO Trajectory Muhammad Aaliyan, Ayesha Imran
    OpenReview
24. #0064
    When Attention Does Not Explain the Peak: Temporal Reference vs. Forecast Output in Attention-Based Time-Series Forecasting Yuji Akamatsu, Takao Yamanaka
    OpenReview
25. #0069
    When the Leaderboard Winner Is Not the Most Grounded: Uncertainty-Aware, Multi-Criteria Evaluation of a Documentation RAG Assistant Evgenii Palnikov, Elizaveta Gavrilova
    OpenReview
26. #0070
    Detecting Contaminated Code-Generation Prompt Batches via Influence Functions Francesco Quinzan, Noor Munir, Yishun Lu, S Roberts
    OpenReview
27. #0077
    Answer-Only Accuracy Measures the Wrong Construct: An Anti-Guessing Verification-Gap Protocol for Trustworthy Olympiad-Math Evaluation Santiago Rivadeneira Quintero, Denys Pushkin, Emmanuel Abbe
    OpenReview
28. #0079
    Is Generative Content Ready to Ship? A Production-Scale Playbook Malay Haldar, Yu Jin, Chin Hui Chew, Hongwei Zhang, Nalini Kartha, Winnie Xiong, Soumyadip Banerjee, Totte Harinen, Aditya Mukherji, Kedar Bellare, Phanindra Ganti, Stephanie Moyerman, Sanjeev Katariya
    OpenReview
29. #0081
    Rethinking Hallucination in Large Language Models: A Legal Framework for Definition, Diagnosis, and Accountability JIA-HONG HUANG, Chao-Chun Yang, Prayag Tiwari
    OpenReview
30. #0085
    It's Not What the Image Shows: Irrelevant Context Destabilises VLM Judges Without Informing Them Nagham Omar, Mahmoud Jabarin, Kinan Ibraheem, Lotem Peled-Cohen
    OpenReview
31. #0086
    Retrieval Rankings Do Not Directly Identify Multimodal Latent Space Alignment Eleonora Grassucci, Danilo Comminiello
    OpenReview
32. #0090
    What Does Fifty Days Stabilize? A Commit-Level Audit of ForecastBench Daniel Alami
    OpenReview
33. #0091
    Contaminated Ancestors: Seven Ways Our Own Evaluation Protocol Nearly Told Us the Wrong Thing Christopher Yu-Hao Huang, Ryan Jaemin Ahn, Wenhao Lu
    OpenReview
34. #0093
    Dissociating Evaluation Awareness from Sandbagging in Open-Weight LLMs (0.5B–32B) Jingting Huang, Zhengran Han
    OpenReview
35. #0095
    BMMS: Policy-Conditioned Measurement of Behavioral Compliance in Language Models Anthony Rhodes, Gadi Singer
    OpenReview
36. #0096
    LLM judges are easier to talk down than up: a Bayesian item response analysis Guoyang Zheng, Susan Wei, Sevvandi Kandanaarachchi
    OpenReview
37. #0097
    Rethinking Vacuity for OOD Detection in Evidential Deep Learning Claire McNamara
    OpenReview
38. #0099
    Frozen LLMs as Recommenders for AutoML Pipeline Selection: Can We Trust the Evaluation? Roman Neruda, Juan Carlos Figueroa Garcia, Carlos Alberto Franco Franco
    OpenReview
39. #0100
    When Checks and Verdicts Disagree: Auditing Aggregation-Contract Fidelity in a Prompt-Structured LLM Judge Family Mingda Yuan, Shu Huang, Tomas Ramanauskas
    OpenReview
40. #0101
    Conditional Evaluation of Language Models with Cheap Auxiliary Signals Zhi Zhang, Lingfeng Lyu, Yue Kang, Doudou Zhou
    OpenReview
41. #0102
    Unreliability or Disagreement? What an Interpercentile Decomposition Can and Cannot Show Pavan Kumar Dubasi
    OpenReview
42. #0103
    Certified Rank-Stability: Risk-Controlling Leaderboard Claims Under Distribution Shift ADITYA UPADHYAY, Arpan Verma, Chandan Sah, Dev Utkarsh Pal, Harshit Gupta
    OpenReview
43. #0108
    Can You Trust a Label-Free Reliability Estimate? An Estimability Threshold for Judge and Detector Panels Under Decision-Induced Censoring Shiva Koreddi
    OpenReview
44. #0109
    Evaluating the Evaluators: Investigating LLM Judges for Personalized Writing Style Assessment Natalie Mackraz, Andrew Silva, Aini Putkonen, Brihi Joshi, Joelle Alcaidinho, Arnav Arora, Barry-John Theobald, Katherine Metcalf
    OpenReview
45. #0110
    Detecting and Mitigating Non-Deterministic Answer Flips for Reliable LLM Inference Qitao Tan, Joymallya Chakraborty, Navaneet K L, Kushal Kumar, Nicolas Dormont, Vidit Bansal, Abhishek Dan
    OpenReview
46. #0111
    When Correctness Is Computable: Auditing AI Evaluators Against an Exact Oracle in Household Finance Vignesh Coumarane
    OpenReview
47. #0112
    Label-free detection and mitigation of judge bias Paul Albert, Quoc Phong Nguyen, Julien Monteil
    OpenReview
48. #0115
    Defining Operational Conditions for Safety-Critical AI-Based Systems from Data Johann Maximilian Christensen, Elena Hoemann, Frank Köster, Sven Hallerbach
    OpenReview
49. #0116
    Does the Proxy Preserve the Decision? A Stopping-Policy Audit of Multiple-Choice Reformulations of Incremental QA Ankit Aggarwal
    OpenReview
50. #0119
    Overwhelmed by Choice: Studying LLM Decision Making at Scale Yu-Chi Lin, Aryan Seth, Anshul Aravind, Eugene Lee, Tanmay Parekh, Nanyun Peng, Kai-Wei Chang
    OpenReview
51. #0120
    “Uncertain” Is Uncertain: Prompt-Induced Abstention in VLMs Is Neither Stable Nor Human-Aligned Fang Hao, Lu-Yan Wang, Shang-Hong Lai
    OpenReview
52. #0121
    Same Output, Different Gold: Measuring How Reference Choice Moves a Multilingual Benchmark Score Parth Kulshreshtha, Shivali Dalmia, Abhishek Mukherji
    OpenReview
53. #0125
    The Evaluation Paradox: Why Capability Evaluations Run Uncontained, and Why Monitoring Does Not Fix It Jaswanth Alkur
    OpenReview
54. #0126
    Voice or Stereotype? Disentangling Acoustic and Content-Based Gender in Speech-to-Speech Models Xiaoqun Liu, Tanu Mitra, Harshit Rajgarhia, Abhishek Mukherji
    OpenReview
55. #0129
    Is the Circuit in the Model or in the Method? Spectral Identifiability for Mechanistic Evaluation Ashim Dhor, Pin-Yu Chen
    OpenReview
56. #0132
    BODYBENCH: Evaluating Adversarial Image Defenses Against AI Nudification Inpainting Li Qiwei, Salma Abdel Magid, Olga Russakovsky, Eric Gilbert, Sarita Schoenebeck
    OpenReview
57. #0133
    Metric Match: A Subset Selection Approach to Evaluating LLM Judge Reliability Alyssa Unell, Natalie Dullerud, Naomi Boneh, Meena Jagadeesan, Tatsunori Hashimoto, Nigam Shah, Sanmi Koyejo
    OpenReview
58. #0134
    Test Accuracy Conflates Learned Weights with Normalization State Sarthak Tayal, Suayb S. Arslan, Emirhan Inan
    OpenReview
59. #0135
    Psychometrics Without a Questionnaire: Auditing an LLM Rater via Item Response Theory (LIRT) Felix Crabtree, Ben Griffin, Yigit Ihlamur
    OpenReview
60. #0137
    Efficient Language Model Evaluation Using Latent-Space Stratification Samuel Myren, Kurtis Shuler, Grant David Meadors, Emily Taylor
    OpenReview
61. #0140
    Can We Trust In-Distribution Success? Locked Evaluation Reveals Transfer Failure and Sampling-Depth Entanglement in CRISPRi Perturbation Prediction Mehrdad Shoeibi, Niloofar Yousefi
    OpenReview
62. #0149
    Evaluating Deception, Not Just Deepfakes Jessee Ho, Shweta Khushu, Shaina Raza
    OpenReview
63. #0150
    Fidelity Ceilings Break Reliability Coefficients in Generative AI Evaluation Mst Mousumi Rizia, Hanna Suominen
    OpenReview
64. #0151
    What You Actually Download: Shipped Defaults Are Mostly Harmless and Occasionally Catastrophic for Browser LLMs Adnan Khan
    OpenReview
65. #0152
    Checking the Output Does Not Check the Aggregator: A Bit-Exact Counterexample in Federated Learning Ruiting Wang
    OpenReview
66. #0155
    Faithful Dual-constrained Erasure for Robust LLM Safety Alignment Jiaqing Li, Shide Zhou, Zhibo Zhang, Yuxi Li, Tianlong Yu, Kailong Wang
    OpenReview
67. #0156
    SIR: Self-improving Red-teaming for Compute Use Agents Chen Xiong, Zhiyuan He, Pin-Yu Chen, Stjepan Picek, Tsung-Yi Ho
    OpenReview
68. #0158
    Same Top Fraction, Different Workload: Auditing Eligibility and Capacity in Clinical Risk Evaluation Vache Oganisyan, Dmitry Lvov, Ilya Pershin
    OpenReview
69. #0161
    Generalization Is Stability, Not Accuracy: Multi-Axis Evaluation of LLMs Nagham Omar, Mahmoud Jabarin, Maya Rozenshtein, Rom Himelstein, Avi Mendelson, Amit LeVi
    OpenReview
70. #0162
    Swap-Consistency Does Not Certify a Code Judge: Separating Calibrated Abstention from Position-Locking with an Execution Oracle Ching-Yu Lin, Yifan Liu
    OpenReview
71. #0164
    Towards a Practical Understanding of On-Policy Lagrangian Safe Reinforcement Learning Lindsay Spoor, Alvaro Serra-Gomez, Aske Plaat, Thomas M. Moerland
    OpenReview
72. #0165
    Are Near-Tied LLM Rankings Robust to Benchmark Composition? Qiaoyuan ZHENG, Yiqu Yang
    OpenReview
73. #0166
    Beyond Accuracy: On Measuring Logical Compliance of Predictive Models Guillaume Delplanque, Pierre Geneves, Nabil Layaïda
    OpenReview
74. #0170
    Is Inter-Seed Cross-Play Enough? Assessing the Standard Evaluation Protocol for Zero-Shot Coordination Maksymilian Wolski, Nicholas Stephen Hoernle, Johannes Forkel, Jakob Nicolaus Foerster
    OpenReview
75. #0171
    Mind the Tokenizer: Why a Single Space Decides What Multiple-Choice Evaluations Measure Prabhu Teja S, Frank Schneider, Martin Reinhardt, Fabien C. Y. Benureau, Thomas F Burns
    OpenReview
76. #0173
    The Screen Decides the Verdict: Operating Characteristics of Model-Exclusion Rules in LLM Evaluation Emile Boullineau, Jose Daniel Muñoz Arciniegas
    OpenReview
77. #0177
    BenchBench-Protocol: Evaluating Real-World Wet-Lab Protocol Reasoning and Modification Aditya Sivakumar, Nithin Parsan, Ashu Singhal, Nicholas Larus-Stone
    OpenReview
78. #0178
    PoisonForge: Task-Level Targeted Poisoning Benchmark for Instruction-Tuned LLMs Luze Sun, Anshuman Suri, Harsh Chaudhari, Cristina Nita-Rotaru, Alina Oprea
    OpenReview
79. #0180
    Do FID Evaluations Bias Generative Models Away from Statistical Relevance? Saumya Goyal, Dhruv Garg, Divyan Goyal, Barnabas Poczos
    OpenReview
80. #0182
    Evolving LLM-Generated Features for Interpretable Classification Jack Butler, Zainab Afolabi, Nikita Kozodoi
    OpenReview
81. #0183
    Can We Trust Evaluations of AI Weather Forecasts for Extremes? How Evaluation Protocols Shape Model Rankings Tobias Biegert, Sam Allen, Zhongwei Zhang, Sebastian Lerch
    OpenReview
82. #0186
    Detection Without Diagnosis in Small Evaluators Rafael Gardos
    OpenReview
83. #0187
    Reproducing ELEPHANT: Judge-Dependence and Model Deprecation in an LLM Sycophancy Benchmark Kristina Sakayeva, Ana Trisovic
    OpenReview
84. #0188
    Stress-Testing Efficient Responsible-AI Evaluation: When Compute Savings Change Benchmark Conclusions Ahmed ELKady, Aravind Narayanan, Yani Ioannou, Shaina Raza
    OpenReview
85. #0192
    Benchmark Contamination: A Taxonomy Organized by Defeated Mitigation Johanna Angulo, Víctor Yeste, Hector Espinos-Morato
    OpenReview
86. #0194
    Training LLMs to Verbalize Evaluation Awareness Usman Anwar, Sahar Abdelnabi, David Krueger
    OpenReview
87. #0198
    Support-Set Target Leakage in Relational Foundation Models during In-Context Learning: Model Dependence and Evaluation Reliability Roshan Reddy Upendra, Alexandre Dorais, Joe Meyer, Andrew Pouret, Anastasios Lambrianos Stappas, Dinesh Katupputhur Ramprasath, Viswanath Ganapathy, Tom Palczewski, Minghua Li
    OpenReview
88. #0202
    PatchTrace Receipts: Auditable Delta Records for Evaluator Maintenance Pengkun Zhang
    OpenReview
89. #0206
    COMPREHEND: Ultra-Fine-Grained Benchmark Understanding by Entailment Hierarchy Sivaraaman Balakrishnan, Prakriti Shankar Shetty, Jeremiah Hong, Hamed Zamani, Haw-Shiuan Chang
    OpenReview
90. #0208
    Architecture-Induced Rank Reversal in Out-of-Distribution Detection Mustakim Ahmed Hasan, Ishtiak Mahmud Saad, Mominul Islam
    OpenReview
91. #0209
    The Benchmark Blind Spot: Lag Profiles of Distributional Model Collapse and a Low-Cost Calibration-Gap Signal Cyrus Hui
    OpenReview
92. #0215
    Beyond Global Shuffles: Auditing the Construct Validity of Clinical Measurement–Text Pairing Evaluations Yaser Kord, Moritz Hess, Jens Timmer
    OpenReview
93. #0217
    Speculative Evaluation of Stochastic LLMs Qianli Shen, Xiang Li, Ruomeng Ding, Yanxi Chen, Daoyuan Chen, Yaliang Li
    OpenReview
94. #0218
    Measuring Stability via Minimal Context Interventions Erel Hadad, Rom Himelstein, Amit LeVi, Brit Youngmann, Avi Mendelson
    OpenReview
95. #0220
    What Does a Passed Evaluator Audit Establish? Observation Sufficiency for Post-Repair Claims Mateo PETEL
    OpenReview
96. #0221
    Beyond Correlation: Validity Conditions for Task-Affinity Evaluation Fangqin Zhou, Joaquin Vanschoren
    OpenReview
97. #0224
    Visual Fairness Misalignment: AI Models Prefer People Based on How They Look Mostufa Jbareen, Amit LeVi, Rom Himelstein, Avi Mendelson, Brit Youngmann
    OpenReview
98. #0225
    EvidenceFlip: Counterfactual Evidence Audits for LLM Judges in Digital Forensics Arash Mahboubi, Yue Wang, Duoyi Zhang, Seyit Camtepe
    OpenReview
99. #0233
    The Errors That Matter: Consequence-Aware Evaluation of LLM Judges Vishal Srivastava
    OpenReview
100. #0234
     Candidate Exposure Changes Abstention: Auditing Construct Validity in Grounded LLM Evaluation Rajeel Ahmad Ansari
     OpenReview
101. #0235
     LLM Judges Verify Presence, Not Absence Dr Sebastian Fox, Luke Markham, Ryan Lail, Michael Karotsieris
     OpenReview
102. #0237
     Fuzzy Evidential Learning: Structured Evidence Generation for Uncertainty-Aware Vision Networks Oliver Pitts, Rossella Arcucci
     OpenReview
103. #0239
     Spline Reorganization Without Grokking: Stress-Testing Knot Entropy as a Progress Measure in KANs RANJAN VEERABHADRASWAMY, Ajith Jubilson E
     OpenReview
104. #0240
     Evaluation Metrics for Safe Reinforcement Learning Lindsay Spoor, Aske Plaat, Thomas M. Moerland
     OpenReview
105. #0247
     When an Evaluation Should Refuse to Answer: A Fail-Closed Weight-Perturbation Pilot in Hebrew Aviad Cohen
     OpenReview
106. #0250
     Patched, Not Repaired: A Low Benign False Positive Rate Certifies the Region, Not the Model Bonam Osene Moto JR, Denis Mugendi
     OpenReview
107. #0251
     Strength Is Not Soundness: Measuring the Exploitability of LLM Policies in Imperfect-Information Games Benjamin Regazzoni, Corentin Caillaud, Rui Miguel Azevedo Leitão
     OpenReview
108. #0267
     The Calibration Trap in Federated Learning Md Ahsan Karim
     OpenReview
109. #0268
     Ask the Model: Counterfactual Cue Reliance Predicts Worst-Group Failure and Targets Repair Prakhar Gupta, Terry Jingchen Zhang, Florent Draye, Bernhard Schölkopf, Zhijing Jin
     OpenReview
110. #0269
     Same Models, Different Winners: How Evaluation Design Changes Model Selection in Financial Volatility Forecasting Vivan Rangra
     OpenReview
111. #0271
     SplitWitness-G: Measuring Train-Visible Evidence in Link-Prediction Benchmarks Vivan Rangra
     OpenReview
112. #0272
     Integrity Is Not Validity: A Layered Assurance Case for Stateful Agent Evaluation Sahand Ghorbanpour, Soumyendu Sarkar
     OpenReview
113. #0278
     DatedGPT: Preventing Lookahead Bias in Large Language Models with Time-Aware Pretraining Yutong Yan, Raphael Tang, Zhenyu Gao, Wenxi Jiang, Yao Lu
     OpenReview
114. #0280
     When Score Contrasts Outrun Their Protocols: Construct Validity in Grounded Generation Alexander J Ryu, Bruce Changlong Xu
     OpenReview
115. #0281
     The Loss Does Not Define the Private Optimizer: Query Semantics in Robust Fine-Tuning Maria Terenteva, Kirill Lukianov
     OpenReview
116. #0283
     A Function-Level Vulnerability Score Measures Flag Rate More Than the Model: Protocol Effects on Paired Benchmarks Maciej Cichoń, Bartłomiej Dmitruk
     OpenReview
117. #0284
     Cited but Not Consulted: A Counterfactual Audit of Legal Chain-of-Thought Faithfulness Saisab Sadhu, Shreeyans Arora, Pratinav Seth
     OpenReview
118. #0286
     What Do Compliance Detectors Read? An Audit of Activation Probes and Guard Models Saisab Sadhu, Aadit Sengupta, Vinay kumar Sankarapu, Pratinav Seth
     OpenReview
119. #0288
     Selective Disclosure of Hidden Directives in Reasoning Models: Behavioral Asymmetry and Steering Simba Shi, Xander Tifft, Wen Xing
     OpenReview
120. #0289
     BILD: A Bilinear Interaction Learnable Discriminator for Post-Hoc Misclassification Detection ANTHONY FERNANDO Judeson, Torsten Schön
     OpenReview
121. #0295
     Guardrail erosion in domain fine-tuned large language models: A paired audit of knowledge-centric abstention in biomedical LLMs Stephen Peng, Vineet Jagadeesan Nair, Noah Broestl, Steven Mills, Anne Friederike Kleppe, Adi Zolotov
     OpenReview
122. #0298
     PyMETA: Evaluating Student Code Diagnosis on and Beyond the First Execution Error Ziqi Tang, Chuyue Li, Jingyi Wang, Yu Wu, Kazuma Hashimoto, Lingyu Gao
     OpenReview
123. #0300
     What Do Evaluation Awareness Metrics Measure Hugh Van Deventer, Laura Ruis, Joshua Joseph
     OpenReview
124. #0301
     The Readout Chooses the Winner: Evaluation Protocols Decide Fairness Conclusions in Continual Learning Keira Chatwin, Miles Bliey
     OpenReview
125. #0302
     The Evaluation Gap: Joint Bounds under Distribution Shift, Selective Labels, and Outcome Measurement Error Sean M. Richardson, Amanda Lee Coston, Ashesh Rambachan
     OpenReview
126. #0305
     Red Teaming is a Measurement Problem: Item Response Theory for Calibrated, Cost-Efficient Robustness Evaluation Sunil Kothari, Rajiv Lal
     OpenReview
127. #0306
     When Evaluation Protocols Outlive Their Assumptions: Stress-Testing Social Desirability Bias Measurement Across LLM Generations Kensuke Okada, Yui Furukawa
     OpenReview
128. #0308
     Truthfulness Is Not Enough: Prompt-Conditioned Evidence Selection in LLMs Eunice Yiu, Sixuan Chen, Joshua Attih
     OpenReview
129. #0309
     Benchmarking Exploration Hacking Propensity in Frontier LLMs Eduard Hübner, Nina Żukowska, Alexander Panfilov, Joschka Braun
     OpenReview
130. #0310
     PriorBPB: Reducing the Answer Length Confound of Bits-per-Byte While Improving Accuracy Prediction Thomas F Burns, Prabhu Teja S, Frank Schneider, Martin Reinhardt
     OpenReview
131. #0311
     VerifierContractBench: Auditing the Safety Validity of Optimized Agent Evaluation Lakshya Narula, Saurabh Tripathi, Lily Bharati Sharma, Nikhil Rao
     OpenReview
132. #0312
     My Graph Doesn’t Help! Analyzing Causal Graph Construction, Representation, and Utility in LLMs Aman Syed, Benjamin Li
     OpenReview
133. #0315
     Where Knowledge and Authority Sit Changes What an Agent Benchmark Can Resolve Dan C. Hsu, Luke Lu
     OpenReview
134. #0316
     Evaluating Agentic AI in Transportation: A Multidimensional Framework and Deterministic Stress Test for Traffic-Simulation Workflows Xiangyong Luo, Wan Li, Guanhao Xu, haowen xu, Abhilasha Jairam Saroj, Jinghui Yuan, Ross Wang
     OpenReview
135. #0317
     When Calibration Records Cannot Support Routing: A Denominator and Artifact Audit Carlos Toxtli Hernandez, Manuel Delaflor
     OpenReview
136. #0318
     Stubborn or Sycophantic? Auditing What Sycophancy Evaluations Measure HanRui Zhang, Ashton Chew, Ryan Soe, Ahmed Taha, Ruizhe Li, Aditya Shah, Maheep Chaudhary
     OpenReview
137. #0321
     Dutch Books for Language Models Isaiah Andrews, Suproteem K Sarkar
     OpenReview
138. #0322
     The Reward Horizon Problem: Learning From Delayed Feedback Eric Fithian, Suproteem K Sarkar
     OpenReview
139. #0323
     Beyond Prediction Error: A Wind Turbine Case Study for CFD Surrogate Model Evaluation Caleb White, Masha Folk
     OpenReview
140. #0324
     Your Benchmark Is an Empirical Measure Over Difficulty Yifan Sun, Naicheng Yu, Jiawen Gong, Jingyan Shen, Huan Zhang
     OpenReview
141. #0326
     How Much Can Repeated Running Buy? A Measured Variance Ceiling for LLM Code Evaluation Yongxi Zhou, Lai Yun Choi, Jiaxi Wen, Wenbo Ye, Junwei Yao
     OpenReview
142. #0327
     When Accuracy Is Not Enough: Evaluating Trustworthiness in High-Stakes Lease Interpretation Bin Liang, Yuxi Lu, Jin Wang, Zhidong Li
     OpenReview
143. #0328
     Which LLM for Which Work? Budgeted Model Allocation under Uncertain Evaluation Hamed Khosravi, Xiaoming Huo
     OpenReview
144. #0329
     CoBe: Counterfactual Benchmark for Text Editing Arvind Raghavan, Inwoo Hwang, Shreyas Havaldar, Kai-Zhan Lee, Aurghya Maiti, Yushu Pan, Jeffrey Wu, Mingxuan Li, Elias Bareinboim
     OpenReview
145. #0332
     A Study of LLM Judge Capability-Dependent Biases and Multi-Judge Ensemble for Bias Calibration Gemma Zhang, Prachi Badarayani, Asmi Kumar, Sadid A. Hasan, Sulaiman Vesal
     OpenReview
146. #0334
     Separating Architecture from Initialization: What a Controlled CNN-Transformer Benchmark Can and Cannot Attribute. Chandra Prakash Bathula
     OpenReview
147. #0337
     ECHO: Diagnosing Spatial Cue Access in Audio-Language Models Daniel Huang, Alvin Shen, Aarush Rachakonda, Aryan Shrivastava, Kyle Xu
     OpenReview
148. #0341
     When the Metric Lies: Silent Evaluation Failures in RL Exploration Seyed Ali Alavi, Mohaddeseh Jahanabadi, Dustin Dannenhauer
     OpenReview
149. #0342
     LLM-Generated Investment Research Memos Need Validation Chenri Luo, Xukang Gong, BINYAN YU, Xiao-Yang Liu
     OpenReview
150. #0343
     Calibration Selection, Not Contamination, Breaks Conformal FDR Control in Graph Anomaly Detection Suhaan Khan, Suhaan Gopal
     OpenReview
151. #0344
     Evaluating Multimodal Benefits in Closed-Loop Robot Control Nihar Mudigonda, Feolu Kolawole
     OpenReview
152. #0346
     SANE: Automated Scorer-Specific Metamorphic Testing for LLM Judges Ethan Ludwick, Yu-Chung Hsiao
     OpenReview
153. #0350
     HOP-CAT: Higher-Order Profile-Constrained Adaptive Testing for Efficient Multidomain LLM Evaluation Chengyuan Yao, Weixuan Xiao, Renzhe Yu
     OpenReview
154. #0352
     Internal Stability Is Not Evidence of Replication: Evaluating Attention-Based Self-Correction Signals Gunjal Gupta, Vidhi Nagda, Nikita Goyal
     OpenReview
155. #0353
     PADMÉ: Preference Alignment Data Synthesis for Meta-Evaluation of LM Agent Evaluators Cheng Chang, Yining Mao, Peng Qi
     OpenReview
156. #0356
     PTXBench: Evaluating LLMs for GPU Kernel Optimization with Architecture-Specific PTX Genghan Zhang, Yixin Dong, Chengze Fan, Zhichen Zeng, Yueming Yuan, Shaowei Zhu, Kunle Olukotun
     OpenReview
157. #0358
     Reading the Room, or Reading the Prompt? Measurement Fragility in Verbalized Evaluation-Awareness Neta Ravid, Daniel Dahl Juhl, Matthew Gregory, Guy Nachshon, Jacob Haimes
     OpenReview
158. #0361
     Auditing Language-to-Action Control in Driving VLAs Aradhya Goel, Bhoomika Gupta, Muhammed Ustaomeroglu, Guannan Qu
     OpenReview
159. #0362
     Asymptotically Log-Optimal Bayes-Assisted Confidence Sequences for Bounded Means Valentin Kilian, Stefano Cortinovis, Francois Caron
     OpenReview
160. #0366
     Same Benchmark, Different Conclusions? Auditing the Stability of Kazakh Large Language Model Evaluation Akylbek Maxutov, Nartay Aikyn
     OpenReview
161. #0367
     Identifying Intelligent Processes via Online Sequential Testing ARITRA DAS, Debayan Gupta
     OpenReview
162. #0368
     Metamorphic auditing of learned AC-OPF evaluation reveals representation dependence Stephen Peng, Vineet Jagadeesan Nair
     OpenReview
163. #0369
     When Pair Count Is Not the Sample Size: What All-Pairs Agent Comparisons Estimate Wei-Jung Huang
     OpenReview
164. #0372
     Model Evaluation for Spatially Structured Data: A Case Study of Satellite Altimetry Surface Reconstruction Xiaoyu Xia, Abani Patra, Beata Csatho
     OpenReview
165. #0375
     Same Scene, Different Answer: Variance and Rollout Budgets for Flow-Policy Evaluation Nihal Gunukula
     OpenReview
166. #0376
     Assessing the Gameability of Vision Language Model Judges for World Model Video Evaluation James Kim, Rahul Bellam, Joseph Le Nevez, Neil Porwal, Vasu Sharma, Nicholas Babey
     OpenReview
167. #0377
     Can Image Captions Support Concept Interpretation? A Controlled Validation Study Farnoosh Javar, Kei Wakabayashi
     OpenReview
168. #0382
     On the estimation and validity of AI time horizons—a statistical look at the METR plot Drew T Nguyen, William Fithian
     OpenReview
169. #0383
     PragAlign: Feedback-Guided Pragmatic Alignment for Controlled Synthetic Dialogue Generation Smitha Muthya Sudheendra, Jaideep Srivastava
     OpenReview
170. #0384
     LLM-as-Environment: Benchmarking Open-Ended Decision Making with Language-Model Simulators Caroline Choi, Neil Band, Nick Rui, Sandra Yang, Zeyneb N. Kaya, Ludwig Schmidt, Tengyu Ma
     OpenReview
171. #0385
     The Camera, the Split, and the Label: A Three-Check Audit of Glaucoma-Progression Prediction Hao Li, Jared Joselowitz, Zachary Ellis, Yuxiang Zhou, Samuel Thio, Aisling Higham, Yajie Vera He
     OpenReview
172. #0386
     Partition, Prompt, Aggregate: Statistical Self-Consistency in Language Models Patrik Wolf, Thomas Kleine Buening, Andreas Krause, Celestine Mendler-Dünner
     OpenReview
173. #0387
     Policy Is Not Capability: Some Safety Benchmarks Measure Which Definition of "Unsafe" You Trained On Vahid Jebraeeli, Jonathon Romero, Yijin Wang, Adrian Chan, Jianbang Zhang
     OpenReview
174. #0389
     Mean Accuracy Is the Wrong Estimand for a Release Decision Jaewon Choi, Jihye Kang, Namhyuk Ahn
     OpenReview
175. #0390
     Which singing technique degrades first? Technique contrast retention for neural audio codecs Yiyang Xu, Zihao Wang
     OpenReview
176. #0391
     Can We Trust What We Measure? Auditing LLM Validity-Awareness in Predictive Modeling Jennifer Zhang, Amanda Lee Coston
     OpenReview
177. #0395
     Drop the Act: Probe-Filtered RL for Faithful Chain-of-Thought Reasoning Swapnil Parekh, Naman Goyal
     OpenReview
178. #0399
     Auditing and Enforcing Calendar-Shift Invariance in Clinical Prediction Models Wenhao Maxwell Ren, Everest Yang, Valarie Tran, Igor Shuryak
     OpenReview
179. #0401
     Testing LLM Ensemble Disagreement as an Ex-Ante Signal of Underspecification in Accounting Standards Qian Deng, Chuanqi Tang
     OpenReview
180. #0402
     Audit the Rule and the Instrument: Two Checks Every Rubric-Evaluation Claim Should Report Tianlang Chen
     OpenReview
181. #0403
     When the Stratifier Is Not Identified: A Pre-Registered Self-Audit of Regime-Stratified Evaluation Tianlang Chen
     OpenReview
182. #0404
     Auditing a Conflicted Judge: Rater-Free Verification of an LLM-Scored Forensic Reliability Benchmark Hemanth Kumar M J, Amogh A S
     OpenReview
183. #0409
     Same Text, Different Prediction: Serving-Context Nondeterminism in Text Classifiers Santhosh Kumar Kasa, Siva Rajesh Kasa, Sumit Negi
     OpenReview
184. #0410
     TVD-Conflict: Disagreement-Guided Discovery of Jailbreak Judge Vulnerabilities Zhouyuan Ma, Pixiao Lan, Yutao Wu
     OpenReview
185. #0411
     When Guardrails Look Effective: Construct Validity Failures in LLM Agent Commerce Evaluation Peiying Zhu, Sidi Chang
     OpenReview
186. #0412
     Same Label, Different Feature: Explainer Dependence and the Causal Validity of SAE Explanations HanRui Zhang
     OpenReview
187. #0413
     Beneath the Scores: Rethinking Hallucination Evaluation for Video Understanding Models Shuzhi Gong, Fengze Sun, Yuansan Liu
     OpenReview
188. #0418
     Stress-Testing RL-based Capability Elicitation with Randomized Underperformance Nina Żukowska, Eduard Hübner, Alexander Panfilov, Joschka Braun
     OpenReview
189. #0423
     Auditing Financial ML Backtests with Evaluation Contracts and Executable Checks Jimmy Chan, Kevin Lee
     OpenReview
190. #0424
     Do LLM Judges Measure What We Mean? Auditing Criterion Alignment and Metadata Dependence in RAG Evaluation Konrad Wojtasik, Marcin Oleksy, Maciej Piasecki
     OpenReview
191. #0428
     SeqPAC: PAC-Certified Labeling with Sequential Ensembles Qianyu Julie Zhu, Ashwin Chandramouli, Youssef Marzouk, Suleiman A. Khan, Fatemeh Mansoori
     OpenReview
192. #0430
     When the Metric Lies: A Full-Stack Audit of a Multi-Agent Evaluation Protocol Jan Safrata
     OpenReview
193. #0444
     Auditing Residualization-Based Evaluation: A Confounding Failure Mode in Cross-Modal fMRI Decoding Jaden Chun, Andy Zhang
     OpenReview
194. #0448
     Change the Metric, Change the Winner: Auditing How We Decide Which Clinical Foundation Model to Deploy in Cancer Pathology Manpreet singh
     OpenReview
195. #0450
     Robust to Which Model Change? A Unified Evaluation of Robust Counterfactual Explanations Marcin Kostrzewa, Maciej Zieba
     OpenReview
196. #0452
     Neither Blinding Nor a Jury, Neither Capability Nor Strategy: What Brings a Verifier to Reliability Johnny Kozman, Arushi Waddepalli, Tilak Parajuli
     OpenReview
197. #0453
     Coordinated, Not Independent: Dependence and Averaging Choices in AI-Prevalence Evaluation Aram Bagdasarian
     OpenReview
198. #0454
     What Does Held-Out Accuracy License? A Stated-Inference Evaluation Protocol for Few-Shot Prompt Tuning Keisuke Yokota
     OpenReview
199. #0455
     When Evaluation Goes Blind: Auditing Patch-Grid Failure Modes in Vision Transformers Steffen Jung, Margret Keuper
     OpenReview
200. #0456
     Auditing VLM Benchmarks from Existing Evaluation Runs Denis Degtiarev, Anna Gromova, Sergey Katkov, Artem Mikhailin, Safina Mukhamedshina, Albert Nasybullin, Karina Shaikhutdinova, Artem Tuzov
     OpenReview
201. #0458
     Selection-Induced Optimism Corrupts Calibration, Not Just Discrimination: A Code-Verified Audit of Hidden-State Hallucination-Detection Pipelines Lakshmi Chakradhar Vijayarao
     OpenReview
202. #0459
     What Do Paired Vision–Language Benchmarks Measure? Cohort Cues, Metric Structure, and a Medical Case Study Ahmed Naser, Sadam Al-Azani
     OpenReview
203. #0461
     Auditing Frozen SAE Readouts as Evaluation Protocols: Spatial Extent Exposes Readout Fragility Krishna Balsara
     OpenReview
204. #0463
     The Off-Switch Failure: When Safety-Repair Evaluation Rewards Model Collapse Pratinav Seth
     OpenReview
205. #0465
     Epistemic Exams Loren K. Fryxell, Kola Ayonrinde
     OpenReview
206. #0469
     Towards Robust, Performance-Based Contamination Detection for LLMs Allenna Tang, Jackie Chi Kit Cheung
     OpenReview
207. #0470
     Short-circuit Error Attribution Fails for Bottleneck Evaluation Tao Gu
     OpenReview
208. #0474
     From Linear Domain Invariance to Deployment Robustness: A Construct-Validity Audit of First-Moment Erasure Yash Tripathi, Arpan Verma
     OpenReview
209. #0475
     Not All Eval-Awareness Is Equal: Capabilities Framing Predicts Compliance Allison Claire Zhuang
     OpenReview
210. #0477
     Mind the Mic Tsz Chai Alan Liu
     OpenReview
211. #0481
     Measuring Definition Sensitivity in LLM Annotation of Social Science Concepts Bogdan Mamaev, Qizhou Wang, Matteo Vergani, Christopher Leckie
     OpenReview
212. #0482
     Beyond Subject-Wise Splits: Auditing Whether Wearable-Health Evaluations Can Support Their Claims Jaden Moon, Yu Yvonne Wu, Andrew Campbell
     OpenReview
213. #0483
     Benchmark Error Tracks the Perturbation, Not the Construct Hyunoh Yeo
     OpenReview
214. #0485
     When Does a Software-Agent Benchmark Measure Engineering Judgment? A Construct and Task-Distribution Audit Sushant Mehta
     OpenReview
215. #0486
     The Retrieval-Robustness Paradox: Data Poisoning in Retrieval-Dependent Domains James Zhou, Evan Chen, Wei Zou
     OpenReview
216. #0487
     The Probability Term in Motion Forecasting Benchmarks Is Not a Calibration Score Jaewoong Yoo, Ganzorig Gankhuyag, Jin-Man Park, Haengseon Son, Kyoungwon Min
     OpenReview
217. #0488
     RankShift: Evaluating Ranking Instability Across Benchmark Dimensions in Document Parsing Halimat Afolabi, Daeyoung Lee
     OpenReview
218. #0498
     Task Informativeness in AI Evaluation Is Objective-Dependent Victoria Kurakata
     OpenReview
219. #0499
     Same Reasoning, Different Words: Testing Linguistic Invariance in Chain-of-Thought Monitoring Joe Smith
     OpenReview
220. #0500
     Can We Trust Small Language Models as Judges? Suchir Salhan, Mingchuan Zou, Paula Buttery
     OpenReview
221. #0501
     Shortcut Sensitivity in Agent-Safety Evaluations Hasan Demirkiran, Jens Ernstberger
     OpenReview
222. #0502
     Spelling Drift Breaks Keyword Refusal Detectors Sarthak Rawat, Vivek Chauhan
     OpenReview
223. #0504
     MemoReason: Evaluating the Effect of Parametric Memory on Contextual Reasoning in LLMs Zineddine Tighidet, Andrea Mogini, Jiali Mei, Patrick Gallinari, Benjamin Piwowarski
     OpenReview
224. #0506
     What Does Multiple Choice Actually Measure? Disentangling Video Understanding from Option-Exploitation Sepand Dyanatkar, Elliott Piercy, Justin Chan
     OpenReview
225. #0513
     What Can Activation Patching Actually Establish? Reachable-State Geometry and the Limits of Mechanistic Evidence Ankush Kadu, Ananth Kalyanasundaram, Aswanth Krishnan
     OpenReview
226. #0514
     When the Label Measures the Estimator: Threshold Leakage in Estimate-Scaled Evaluation Labels Arush Gumber, Sidhant Gumber
     OpenReview

No papers match your search.