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
252 results for “keyword:"agent eval"” in ICLR 2027. 49 withdrawn or desk rejected listed apart. clearsort:relevanceratingnewestmost read
[PDF]Finding the Move Is Not Winning the Game: XiangqiBench for Closed-Loop Evaluation of LLM Agents▼1361.33
[PDF]Beyond Execution: Operationality and Scientific Fidelity Are Distinct Capabilities of Scientific Coding Agents▼1611.50
[PDF]Self-Play Evaluation Under Population Exposure Mismatch: Unidentified Verdicts on Defensive Skill in Competitive Mahjong▼4.3k2.00
[PDF]Disentangling Agent Contribution in LLM Multi-Agent Systems through Controlled Interventions▲5.4k3.00
[PDF]The Emperor's New Scaffold: Semantically Equivalent Harnesses Reshuffle the Leaderboard▼4.3k2.00
[PDF]Repeatable but Not Replicable: Evaluator State as a Hidden Facet in Self-Evolving Agents▼4.3k2.00
[PDF]When Does a Memory Descendant Need Repair? Structural Lineage vs. Stale State in Persistent Agent Memory▲452.75
[PDF]Generalization Beyond Coding: An Empirical Comparison of Coding Agents and Purpose-Built Agents▲5103.00
[PDF]ALAMEINBENCH: EVALUATING LONG-HORIZON STRATEGIC PLANNING OF LLM AGENTS IN A HISTORICAL WARGAME▲13k2.33
[PDF]SILICA-BENCH: EVALUATING LONG-HORIZON LLM AGENTS THROUGH VERIFIED RTL-TO-GDS IMPROVEMENT▲2843.00
[PDF]Should, Say, Do: Evaluating Moral Consistency from Judgment to Tool Action in LLM Agents▼5.8k2.33
[PDF]The Sensing Gap: Counterfactual Trajectory Forks for Contract-Relative Pre-Commitment Sensing▼1361.33
[PDF]A Decade in a Day: Synthesizing Agent-Native Interfaces from Human-Oriented Software Evolution▼5.9k2.50
[PDF]Out of Step with Evidence: Task Progress State Mismatch in Long-Horizon Tool-Use Agents▼1.2k1.67
[PDF]TIMELINE-BENCH: Evaluating Agents on Realistic Video-Editing Tasks, from Raw Footage to Final Cut▼5.8k2.33
[PDF]Demonstration Conditioning Generally Improves Jagged Frontier Model Performance on Digital and Physical Agentic Tasks▼71.25
[PDF]SPINE: Structured Policy Refinement with Intent-to-Execution Lineage for LLM Software Agents▼4.3k2.00
[PDF]Causal-Discovery-Bench: A Verifiable Sandbox for Diagnosing and Improving LLM Agents in Sequential Causal Discovery▲3972.67
Withdrawn or desk rejected (49)
[PDF]SWE-Climb: Benchmarking Coding Agents on Real Work in Million-Line Brownfield CodebasesWithdrawn1.33
[PDF]The Serialization Bottleneck: Separating State Preservation from Property Derivation in Tool-Using AgentsWithdrawn1.67