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
251 results for “keyword:"agent evaluation"” in ICLR 2027. 49 withdrawn or desk rejected listed apart. clearsort:relevanceratingnewestmost read
[PDF]Discovery or Memorization? A Faithful Benchmark for Real-World Business-Logic Vulnerability Discovery▲5103.00
[PDF]When Tools Silently Lie: Evaluating and Mitigating Blind Compliance in Tool-Augmented Data Agents▲6.1k2.00
[PDF]Deviation Is Not Failure: Decoupling Plan-Adherence from Task-Correctness in LLM Judges via Input-Isolated Single-Axis Fine-Tuning▲6.1k2.00
[PDF]Is This Agent Evaluable? Certifying Verdicts About LLM Agents Under Nondeterminism, Oracle and Coverage Obstructions▼1.2k1.67
[PDF]Where Documentation Ends: Source-Derived Behavioral Specifications for LLM Tool Simulation▼3.7k2.25
[PDF]DEEPSEARCHDAG: A DAG-STRUCTURED BENCHMARK FOR NODE-LEVEL EVALUATION OF DEEP SEARCH AGENTS▼1361.33
[PDF]Reported Safe, Blind in Practice: A Cross-Modal Conflict Benchmark for the Action Layer of Multimodal Agents▼791.00
[PDF]Where Did the Agent Go Wrong? Localizing—and Fixing—Failure Steps in Multi-Step Video-Editing Agent Traces▼5.8k2.33
[PDF]TraceDance: An Automated System for Building Agent Behavior Benchmarks from Real-World Agent Deployment Traces▼1.2k3.00
[PDF]Logically Redundant, Behaviorally Different: Omitting Derivable Information in LLM Agents▼4.3k2.00
[PDF]Measuring Say-Do Consistency in Tool-Using Agents: A Claim-Execution Traceability Framework▼1.2k1.67
[PDF]What Does Context Compression Cost an Agent? Interaction Costs Unrevealed by Task-Completion Metrics▼1361.33
[PDF]DataResearchBench: Benchmarking Visual Data Sourcing with Generalist Agents on the Open Web▼5.8k2.33
[PDF]What Can a Leaderboard Certify? Compositional Controllability for Fair Evaluation and Training of Biomedical Literature-Review Agents▼1.3k1.80
[PDF]Uncovering Robustness Weaknesses in LLM Agents via Persona Evolution with Task-Integrity Constraints▼4.5k2.67
[PDF]ToolPlan-Bench: Diagnosing Hidden Planning Failures in Long-Horizon Tool-Calling Agents Beyond Success Rate▲3.9k1.75
Withdrawn or desk rejected (49)
[PDF]The Repair Is Not the Task: Execution Records and False Completion in Agent RecoveryWithdrawn1.33
[PDF]EarthArena: Benchmarking Agents on Real-world, Long-Horizon Scientific Tasks in Earth ObservationWithdrawn1.67
[PDF]AskInfer-Bench: Profiles Are Evidence, Not Intent Labels, in Personalized Deep ResearchWithdrawn1.00