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
819 results for “keyword:"reward"” in ICLR 2027. 154 withdrawn or desk rejected listed apart. clearsort:relevanceratingnewestmost read
[PDF]The Protocol Is the Agent: Decomposing RL, SFT, and Scaffold Effects in Tool-Use Agents from 0.8B to 27B▼1.2k1.67
[PDF]Time and Cause: Reliability-Adaptive Hybrid Credit Assignment in Reinforcement Learning▲9.2k2.67
[PDF]Environmental Feedback Modeling Matters: Rethinking Feedback Treatment in Agentic Hindsight Self-Distillation▼4.3k2.00
[PDF]Ground the Past to Predict the Future: Optimizing Natural-Language User Representations Without Downstream Supervision▲5.4k3.00
[PDF]Scaling Feedback Alone Is Not Scaling Supervision: From Prediction Quality to Decision Quality in Reward Learning▼1361.33
[PDF]RACS: Response Activation Consistency-guided Data Selection for Reinforcement Learning with Verifiable Rewards▼3.7k2.25
[PDF]Urban-Centric Latent Multi-Task Offline Reinforcement Learning with Mask-Empowered Reward Learning▲5.4k3.00
[PDF]PaRa-OPD: Pathology Reward-Aware On-Policy Distillation via Token-Level Supervision Reweighting▼4.3k2.00
[PDF]Align as Act: Step-level Credit Assignment for LLM Agents via Sequential Subspace Residuals▲3.0k2.33
[PDF]NF-Bridge: Faithfully Bridging Natural and Formal Languages for Mathematical Reasoning▼1.3k1.75
Withdrawn or desk rejected (154)
[PDF]Auditing Reward-Model Selection on Tool Trajectories: Identification, Policy Utility, and Natural TransferWithdrawn1.67
[PDF]Verifier-Induced Strategy Collapse: Family-Correlated False Negatives Eliminate Valid Strategies While Capability Metrics ImproveDesk rejected
[PDF]Reflection Rewards Pay for Deferral: Capability-Aware Rewards for Reflective RLDesk rejected3.00