RLHarness: Co-evolving Procedural Skills with Reinforcement Learning for Long-horizon Multimodal Reasoning

arXiv:2609.32326 · cs.AI · Submitted 2026-09-26 · Read on arXiv

cs.AI

Submitted: 2026-09-26

Updated: 2026-09-26

Code: https://github.com/Ziqiao-Shang/RLHarness

Terminology

• Verifier — 41× in this paper · explained in 2 episodes
• Trace — 33× in this paper · explained in 2 episodes
• Ablation — 26× in this paper · explained in 1 episode
• Distillation — 25× in this paper · explained in 2 episodes
• Accuracy — 16× in this paper · explained in 1 episode
• Adaptation — 16× in this paper · explained in 1 episode
• Reinforcement Learning — 11× in this paper · explained in 4 episodes
• Verification — 11× in this paper · explained in 2 episodes
• Multimodal Reasoning — 10× in this paper · explained in 5 episodes
• Benchmark — 9× in this paper · explained in 2 episodes
• Consistency — 8× in this paper · explained in 3 episodes
• F1 Score — 4× in this paper · explained in 3 episodes
• GRPO — 4× in this paper · explained in 6 episodes
• Agent Skills — 3× in this paper · explained in 1 episode
• Alignment — 3× in this paper · explained in 1 episode
• Aggregation — 2× in this paper · explained in 1 episode
• Grounding — 2× in this paper · explained in 3 episodes
• Pruning — 2× in this paper · explained in 4 episodes
• alpha — 2× in this paper · explained in 1 episode
• AI Agents — 1× in this paper · explained in 2 episodes
• Actor — 1× in this paper · explained in 3 episodes
• Agentic Reasoning — 1× in this paper · explained in 1 episode
• Agentic Workflows — 1× in this paper · explained in 1 episode
• Benchmarking — 1× in this paper · explained in 3 episodes
• Conference on Language Modeling — 1× in this paper
• Credit assignment — 1× in this paper · explained in 3 episodes
• Failure Modes — 1× in this paper · explained in 3 episodes
• Generalization — 1× in this paper · explained in 13 episodes
• Generative AI — 1× in this paper · explained in 2 episodes
• Procedural Knowledge — 1× in this paper · explained in 2 episodes
• Provenance — 1× in this paper · explained in 4 episodes
• Verifiable Rewards — 1× in this paper · explained in 2 episodes
• Visual Grounding — 1× in this paper · explained in 2 episodes

Sources

• HarnessForge: Joint Harness and Policy Evolution for Adaptive Agent Systems
• Co-Harness: Co-Evolving Harnesses and Model Weights for LLM Agents
• Self-Play Meets Skill Evolution: Self-Evolving Search Agents that Pose, Solve, and Remember
• SWE-Skills-Bench: Do Agent Skills Actually Help in Real-World Software Engineering?
• ReSkill: Reconciling Skill Creation with Policy Optimization in Agentic RL
• Skill Self-Play: Pushing the Frontier of LLM Capability with Co-Evolving Skills
• MemoHarness: Agent Harnesses That Learn from Experience
• Prioritizing the Best: Incentivizing Reliable Multimodal Reasoning by Rewarding Beyond Answer Correctness
• WHALE: A Simple Recipe for Joint Harness-Weight Optimization
• Meta-Harness: End-to-End Optimization of Model Harnesses
• SkillsBench: Benchmarking How Well Agent Skills Work Across Diverse Tasks
• SkillGraph: Skill-Augmented Reinforcement Learning for Agents via Evolving Skill Graphs
• ARISE: Agent Reasoning with Intrinsic Skill Evolution in Hierarchical Reinforcement Learning
• Anticipatory Planning for Multimodal AI Agents
• MUSE-Autoskill: Self-Evolving Agents via Skill Creation, Memory, Management, and Evaluation
• Agentic Harness Engineering: Observability-Driven Automatic Evolution of Coding-Agent Harnesses
• Adaptive Auto-Harness: Sustained Self-Improvement for Agentic System Deployment on Open-Ended Task Streams
• SafeEvolve: Harness-Policy Co-Evolution from Agent Experience for Safety Alignment
• EvoHarness-RL: Learning Runtime Harness Coordination for Self-Evolving Agents

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
• 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
• VSAL: A Vision Solver with Adaptive Layouts for Graph Property Detection