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
69 results for “keyword:"agentic rl"” in ICLR 2027. 11 withdrawn or desk rejected listed apart. clearsort:relevanceratingnewestmost read
[PDF]Rewarding Beliefs, Not Only Actions: Belief-Consistent Reinforcement Learning for Long-Horizon Agents▼1.2k1.67
[PDF]Discover and Internalize: A Continual Co-Evolving Training Loop for Self-Improvement Agents▼1.2k1.67
[PDF]Verifiable Hidden Dynamics Play: Generating Agentic RL Environments from Solved Mechanisms▼5.9k2.50
[PDF]MetaAgent-X : Breaking the Ceiling of Automatic Multi-Agent Systems via End-to-End Reinforcement Learning▼4.3k2.00
[PDF]Thinking Beyond Videos: Unifying Video Reasoning and Deep Research for Open-World Video Agents▼5.1k2.60
Withdrawn or desk rejected (11)
[PDF]StepWeb: Trajectory-Conserving Step-Level Credit for Multi-Turn Web RefinementDesk rejected2.00
[PDF]Beyond Timestamps: Decision-Aligned On-Policy Distillation for Long-Horizon AgentsWithdrawn1.67
[PDF]REMIND: Learning Evidence-Conditioned Memory Invocation over Long-Term ConversationsWithdrawn1.33
[PDF]ARBOR: Branch-Structured Credit Assignment for Reinforcement Learning with Verifiable RewardsDesk rejected
[PDF]Dual-Axis Policy Optimization for LLM Agents: Bayesian Feedback Attribution and Trajectory Mass NormalizationDesk rejected2.75