SkillScaffold: Scaffolding Agentic Reinforcement Learning with Policy-Adaptive Skill Credit Assignment and Selective Injection

arXiv:2610.18807 · cs.LG, cs.CL · Submitted 2026-10-13 · Read on arXiv

cs.LG, cs.CL

Submitted: 2026-10-13

Updated: 2026-10-13

Code: https://github.com/Vaidurya00/SkillScaffold

License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Terminology

• Credit assignment — 1× in this paper · explained in 3 episodes
• Reinforcement Learning — 1× in this paper · explained in 4 episodes
• Scaffolding — 1× in this paper · explained in 3 episodes

Sources

• Mem0: Building Production-Ready AI Agents with Scalable Long-Term Memory
• Agent Skills Can Be Harmful: An Empirical Study of Skill-Induced Failures in LLM Agents
• Evolving-RL: End-to-End Optimization of Experience-Driven Self-Evolving Capability within Agents
• From Self-Evolving Synthetic Data to Verifiable-Reward RL: Post-Training Multi-turn Interactive Tool-Using Agents
• ReSkill: Reconciling Skill Creation with Policy Optimization in Agentic RL
• SKILL0: In-Context Agentic Reinforcement Learning for Skill Internalization
• Skill Set Optimization: Reinforcing Language Model Behavior via Transferable Skills
• DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models
• Dynamic Skill Lifecycle Management for Agentic Reinforcement Learning
• Skill1: Unified Evolution of Skill-Augmented Agents via Reinforcement Learning
• ALFWorld: Aligning Text and Embodied Environments for Interactive Learning
• Skill-SD: Skill-Conditioned Self-Distillation for Multi-turn LLM Agents
• Reinforcement Learning for Self-Improving Agent with Skill Library
• ScienceWorld: Is your Agent Smarter than a 5th Grader?
• Agent Workflow Memory
• SEED: Self-Evolving On-Policy Distillation for Agentic Reinforcement Learning
• SkillRL: Evolving Agents via Recursive Skill-Augmented Reinforcement Learning
• ReAct: Synergizing Reasoning and Acting in Language Models
• Co-Evolving Skill Generation and Policy Optimization
• Agentic Policy Optimization via Instruction-Policy Co-Evolution

Related papers

• Polynomial-Augmented Neural Networks (PANNs) with Weak Orthogonality Constraints for Enhanced Function and PDE Approximation
• AIRL-S: Unifying Reinforcement Learning and Search-Based Test-Time Scaling via Adversarial Inverse Reinforcement Learning
• Transformers as Bayesian In-Context Experimenters: Smoothness-Adaptive Efficient ATE Estimation
• Convergence issues in Relational Concept Analysis based on AOC-posets
• Beliefs Beyond Posteriors: The Predictive Cost of ELBO Binding
• Understanding Diffusion Models via Ratio-Based Function Approximation with SignReLU Networks