Reward Design for RL

RL: Reinforcement Learning

Overview

Reward design studies how to turn task goals, preferences, and constraints into learning signals, especially when success is sparse or objectives compete. Approaches include potential-based shaping and reward machines, preference and rubric models, visual progress rewards, intrinsic exploration bonuses, and automated reward search using language models. These methods are used in settings from robot manipulation and locomotion to language-model reasoning, tool use, and multi-agent coordination. A central tension is that denser or adaptive rewards can improve credit assignment but also mislead policies, destabilize training, or invite reward hacking, so evaluation must test intended outcomes rather than proxy scores alone.

Latest papers

CardsList

289 papers37 papers in the last four weeks, down 16% on the four weeks before. 0.3% of all new papers.

  1. Honesty Is a Race: Predicting When Verifiable-Reward RL Teaches a Model to Bluff

    Nov 11, 2026Nathaniel Bottman, Alexei OblomkovReinforcement Learning with Verifiable RewardsLLM Honesty

  2. Towards Better Legal Reasoning LLMs: Signal Balancing and Reward Scheduling

    Nov 11, 2026Yinghao Hu, Qian Xiao, Yuting Huang +7Legal ReasoningStructured Reasoning in Language Models

  3. Temporal Behavior Trees for Reinforcement Learning: Specifications as Rewards, Curricula, and Diagnostics

    Nov 10, 2026Vladimir Krsmanovic, Bernd FinkbeinerReinforcement LearningFormal Methods for RL

  4. SLA-Aware Task Offloading in the IoT Edge-Cloud Continuum: Class-Weighted Deep Reinforcement Learning Against an Exact Optimum

    Nov 7, 2026Shilpa Golla, Tharun Kumar Korine Palli, Urmila BandaruComputation OffloadingReinforcement Learning

  5. Reward Budgeting Reduces Premature Convergence in Reinforcement Learning for LLM Reasoning

    Nov 4, 2026Mengni Jia, Mengyu Zhou, Xiaoxi Jiang +1RL for Language Model ReasoningReward Design for RL

  6. RIDGE: State-Conditioned Reward Blending for Behavioral Coverage in Deep RL Game Agents

    Nov 3, 2026AL Muqshith Mohammed Shifan, Kevin Christopher Chua, Ali Neshati +2Behavioral Diversity in RLReward Design for RL

  7. When Rewards Depend on the Group, GRPO Plays a Tournament

    Nov 2, 2026Junyi Gao, Yu Shi, Pingzhao Hu +1Group Relative Policy OptimizationReward Design for RL

  8. Rewarded for Not Getting Caught: Why Alignment Training Selects for Conditional Compliance and Cannot Instill Norms

    Nov 1, 2026Kevin Baum, Ruta Binkytė, Gerrit Großmann +2AI AlignmentLLM Safety Alignment

  9. Silent Reward and Environment Defects in EV Charging Simulation Benchmarks: A Diagnostic Checklist

    Nov 1, 2026Nour-Eddine Moumni, Rachid Alaoui, Driss KiouachRL BenchmarksSimulation-Based RL

  10. Alignment Flywheel-Guided Suppression Patching for Interactionless Inverse Reinforcement Learning

    Oct 31, 2026Elias Malomgré, Pieter SimoensReward Design for RLReward Hacking

  11. When Does Exploration Prevent Collapse? Three Regimes and the Path of the Reward in Self-Consistency Training

    Oct 31, 2026Jaeho Hwang, Uiwon HwangTest-Time RLRL Exploration

  12. Reward-Adaptive Co-Evolution of Policy and Rubric

    Oct 29, 2026Yizhu Chen, Jianguang Jiang, Jiameng Huang +5Rubric-Based RLReward Design for RL

  13. Event2Reward: Learning Robotic Rewards from Progress Relations

    Oct 29, 2026Junnan Nie, Zitian Fan, Jiachen Zhang +14Robotic RLReward Design for RL

  14. GARDEN: Generalist Autonomous Reward Design Through Evidence Navigation for Cooperative Multi-Agent Reinforcement Learning

    Oct 27, 2026Dogan UrgunReward Design for RLCooperative MARL

  15. Repulsive Population Policy Gradients for Behaviorally Diverse Reinforcement Learning

    Oct 27, 2026Saeed Hedayatian, Harsh Shah, Celine Mneimneh +1Reinforcement LearningBehavioral Diversity in RL

  16. Audit Where Hacks Pay: Budgeted Strong-Verifier Audits Against Reward Hacking in GRPO

    Oct 26, 2026Shurun Li, Michael Hua Wang, Haibo Zeng +1Reward HackingReinforcement Learning with Verifiable Rewards

  17. Improving Reasoning with Rollout Agreement for Group-Based Policy Optimization

    Oct 26, 2026Sohyun Son, Yerim Hong, Jaeyun Park +1Reinforcement LearningRL for Language Model Reasoning

  18. From Human Intent to In-Orbit Action: A Self-Learning Large Language Model Agent for 6G Satellite Networks

    Oct 25, 2026Zhengqian Xu, Yasheng Zhang, Chenhua Sun +2LEO Satellite Networks6G Networks

  19. Eat Your Precision: Precision Objective Mitigates Reward Hacking in Rubric-based RL

    Oct 24, 2026Hyun Ryu, Jong-Heon Kim, Yiqing Xie +3Rubric-Based RLReward Design for RL