NeurIPS 2026 reviews are up: reviews, scores, rebuttals and meta-reviews for 7,877 papers. See the highest rated
acceptodds: the papers of ICLR 2027
37 results for “keyword:"benchmark validity"” in ICLR 2027. 9 withdrawn or desk rejected listed apart. clearsort:relevanceratingnewestmost read
[PDF]The Selection Rule Decides the Winner: A Pre-Registered Audit of Open-Set Graph Anomaly Detection▼4.3k2.00
[PDF]Belief Elicitation Is an Intervention: Measuring and Manufacturing Belief–Action Consistency in Language Models▲3.0k2.33
[PDF]Fails Before Repair Is Not Enough: Auditing the Validity of Executable Repair Benchmarks.▼1351.33
[PDF]When Aggregate AUROC Ranks the Shortcut: Trivial-Axis Headroom in Vision-Language Hallucination Detection▲5103.00
[PDF]The Wrong Kind of Benchmark Difficulty: Diagnosing Benchmark Errors Using Model Disagreement▲3.0k2.33
[PDF]Evaluating Reasoning under Multimodal Conflict: Separating Score Gains from Evidence Selection▼1351.33
[PDF]Candidate Order in LLM-Generated Agent Environments: Prompted Placement, Measured Difficulty and Learned Reliance▼4.3k2.00
[PDF]Can the Simulated Student Learn? Fitted Knowledge-Tracing Models Disagree on Spacing, Recency and Transfer▲11k2.00
[PDF]Compression Is Not Evaluation-Neutral: Fixed RAG Compression Can Distort Reader Comparisons▼1351.33
[PDF]Counterfactual Twins and Executable Validity Gates for Diagnosing Spatial Reasoning in Multimodal LLMs▲11.00
[PDF]CONSTRUCTING A PROVENANCE-ATTRIBUTION BENCHMARK OVER THE SCIENTIFIC LITERATURE: A SUPPLY CENSUS AND A SHORTCUT GATE▼1601.50
[PDF]When Success Is a Lossy Record: Matched-Success Continuation Evaluation for Persistent Agents▼4.3k2.00
[PDF]From Scores to Claims: A Validity Argument Framework for Large Language Model Evaluation▼1.2k1.67
[PDF]Correct Scores, Unsupported Conclusions: Benchmarking Evaluation under Incomplete Semantic Reference▼4.3k2.00
[PDF]When Not to Trust a Global Gaussian Process Posterior: Local Scoring and Progressive Dissonance for Multi-Modal Scientific Discovery▼1.2k1.67
[PDF]One Capability or Many? Structural and Predictive Tests of Benchmark Validity Disagree About Economic Benchmarks for Frontier AI▼71.25
[PDF]TRUSTCONS: EXACTLY VERIFYING CONSISTENCY, COMPLETENESS , AND CONDITIONED RE-ENUMERATION IN LANGUAGE MODELS▼1351.33
[PDF]Judging by the Cover: Cleaning LLM Truthfulness Benchmarks to Avoid Surface-Level Feature Leakage▼5.8k2.33
[PDF]Erase It Where You Found It: Matched-Position Erasure Lowers Demographic Dependence at a Measured Cost▼4.3k2.00
Withdrawn or desk rejected (9)
[PDF]When Is a Policy Violation a Model Failure? Policy Observability and Failure Attribution in Agent-Safety EvaluationDesk rejected
[PDF]What a 3D Reconstruction Benchmark Can and Cannot Tell About Geometric Foundation ModelsWithdrawn2.00
[PDF]When Evidence Changes the Subject: Subject-Typed Claim Licensing for Learned RoutingWithdrawn1.67
[PDF]Learning to Be Fooled: pretraining grows knowledge but keeps the scar of framing susceptibilityDesk rejected
[PDF]Can We Make an LLM Dumber, or Is It Just Playing Dumb? An Empirical Study of Believable Performance ThrottlingWithdrawn1.67