Beyond Binary Verifiable Rewards: Behavioral Self-Rewarding for SWE Agents
cs.AI, cs.SE
Submitted: 2026-10-08
Updated: 2026-10-08
Terminology
Sources
- Reinforcement Learning for Long-Horizon Interactive LLM Agents
- CWM: An Open-Weights LLM for Research on Code Generation with World Models
- Training Long-Context, Multi-Turn Software Engineering Agents with Reinforcement Learning
- DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning
- SWE-TRACE: Optimizing Long-Horizon SWE Agents Through Rubric Process Reward Models and Heuristic Test-Time Scaling
- R2E-Gym: Procedural Environments and Hybrid Verifiers for Scaling Open-Weights SWE Agents
- Agent Lightning: Train ANY AI Agents with Reinforcement Learning
- Training Software Engineering Agents and Verifiers with SWE-Gym
- Improving Ranking Quality and Fairness in Swiss-System Chess Tournaments
- Proximal Policy Optimization Algorithms
- DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models
- PAPO: Stabilizing Rubric Integration Training via Decoupled Advantage Normalization
- Kimi K2.5: Visual Agentic Intelligence
- RAGEN: Understanding Self-Evolution in LLM Agents via Multi-Turn Reinforcement Learning
- Toward Training Superintelligent Software Agents through Self-Play SWE-RL
- Agentless: Demystifying LLM-based Software Engineering Agents
- Self-Rewarding Language Models
- Group Sequence Policy Optimization
Related papers
- MAVEN-T: Reinforced Heterogeneous Distillation for Real-Time Multi-Agent Trajectory Prediction
- Model Discovery Agent: LLM-assisted Bayesian experiment design for data-efficient discovery of mechanistic world models
- Agents' Last Exam
- The Clinician's Veto: Navigating Trust, Liability, and Uncertainty in Autonomous AI Prescribing
- MindHelper: Closed-Loop Embodied Mental-State Reasoning for Precision Intervention
- Incumbent Advantage: Brand Bias and Cognitive Manipulation Dynamics in LLM Recommendation Systems