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819 results for “keyword:"reward"” in ICLR 2027. 154 withdrawn or desk rejected listed apart.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]What Does a Failed Rollout Supervise? Separating Policy Failure from Recoverability in Tool Agents▼4.3k2.00
[PDF]Proper Scoring under Group Reward Normalization: Target Distortion and Exact Finite-Group Updates▼4.3k2.00
[PDF]Let the Weak Judge: Annotation-Free Process Rewards for Multimodal E-commerce Relevance▼1.2k1.67
[PDF]Layout Beyond Format: Internalizing Spatial Execution in LLMs via Element-Level Geometric Rewards▼4.3k2.00
[PDF]Dual-Agent Co-Design of Terrain Curricula and Reward Functions for Mixed-Terrain Quadrupedal Locomotion▼4.3k2.00
[PDF]From Trajectories to Grounded Preferences: Process Preference Synthesis via Interaction Element Graphs for Web PRMs▲3.0k2.33
[PDF]Learn What’s Left, Not What’s Mastered: Saturation Aware Advantage Reweighting for Multi-Reward Policy Optimization▼5.8k2.33
[PDF]RHContrast: A Matched-Control Benchmark for Evaluating Reward-Hacking Risks in Long-Form LLM Responses▲3.0k2.33
[PDF]Transfer One to All: Anchor Query-Specific Rubric Set Learning with Coarse Preferences▼4.3k2.00
[PDF]DESIGN THEFEEDBACK, NOTJUST THEREWARD: ENVIRONMENTS FOROPEN-ENDED REINFORCEMENT LEARNING▼4.3k2.00
[PDF]Right Answers for the Right Reasons? Bridging Reasoning Utility and Visual Dependency in Multimodal Reinforcement Learning▲3952.67
[PDF]PULSE: Provenance-Aware User-Profile Learning and Selective Evaluation for Long-Term LLM Personalization▲3952.67
[PDF]Reproducing, Analyzing, and Detecting Reward Hacking in Rubric‑Based Reinforcement Learning▲3952.67
[PDF]When Return Bounds Do Not Control Update Weights: Prediction Rewards in Multi-Turn LLM Agents▼5.9k2.50
[PDF]LineupRL: Verifiable Reinforcement Learning for Time Series Captioning via Caption-to-Series Identification▲3952.67
[PDF]Beyond Policy Rollouts: Embedding-Space Directional Exploration with Anchored Noise for LLM RLVR▼4.5k2.67
[PDF]LIAR: Length-Interval-Based Reward Shaping for Reliable and Accurate Chain-of-Thought Reasoning▲5.4k3.00
[PDF]Quality Determines Direction, Length Shapes Magnitude: Length Control for Open-Ended Reinforcement Learning▲3952.67
Withdrawn or desk rejected (154)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