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819 results for “keyword:"reward"” in ICLR 2027. 154 withdrawn or desk rejected listed apart.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]Judging in Latent Space: Efficient Generative Reward Modeling via Semantics-Preserving Compression▲6.1k2.00
[PDF]From-Base RLVR for Verify-Only Judges: A Rule-Based Reward Turns Base Models into Trajectory Auditors▲11.00
[PDF]EPIC: Rethinking Trust Regions for Autoregressive LLM Policies via Exact-KL Prefix Importance-ratio Clipping▼1.2k1.67
[PDF]Rewarding the Graph Behind the Chain: Topology-Aware Process Supervision for RL and Reasoning Distillation▲3952.67
[PDF]Pressure Is Not Enough: A Causal Test of Whether Optimization Induces Encoded Chain-of-Thought▼1.2k1.67
[PDF]Benchmarking and Training Reward Models for Cross-Source Evidence Grounding in Search Agents▼3.7k2.25
[PDF]From Rewarding What Models Say to How They Think: Attention Process Reward Model for LLM Reinforcement Learning▼4.3k2.00
[PDF]Improving the Credit Assignment of Group Relative Policy Optimization via λ-Return Advantage Estimation for LLM Reasoning▼1.2k1.67
[PDF]Sufficiency of Zeroth-Order Reward Shaping for Policy Gradient in Stabilization Control▼1.2k1.67
[PDF]GeoVR: Structured Execution Rewards for Reinforcement Learning in Geospatial Code Generation▲3.0k2.33
[PDF]CommerceVibe: Learning to Design E-Commerce Creatives as Executable Visual Code via Dual-Feedback Reinforcement Learning▼5.8k2.33
[PDF]Different Bonuses, Similar Outcomes: Understanding Auxiliary Reward Non-Separation in PPO▼1361.33
[PDF]Small Steps on the Geodesic: Ability-Anchored Progressive Curricula for RL on Reasoning▼5.8k2.33
[PDF]SWE-MILE: Asynchronous Potential-Induced Milestone Credit Assignment for Long-Horizon Software Engineering Agents▼5.9k2.50
[PDF]From News to Explainable Forecast: Commonsense, Causal, and Logic-Consistent Reflection for LLM Forecasting▲5.4k3.00
[PDF]Boosting Reinforcement Learning with Verifiable Rewards via Randomly Selected Few-Shot Guidance▲5103.00
Withdrawn or desk rejected (154)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