Dynamic Skill Lifecycle Management for Agentic Reinforcement Learning

arXiv:2605.10923 · cs.LG, cs.CL · Submitted 2026-05-11 · Read on arXiv

cs.LG, cs.CL

Submitted: 2026-05-11

Updated: 2026-10-10

Comments: Implementation code is available at https://github.com/ejhshen/SLIM

Code: https://github.com/ejhshen/SLIM

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

Terminology

• Reinforcement Learning — 2× in this paper · explained in 4 episodes
• Agentic Reinforcement Learning (RL) — 1× in this paper
• Reinforcement Learning (RL) — 1× in this paper · explained in 25 episodes

Sources

• Back to Basics: Revisiting REINFORCE Style Optimization for Learning from Human Feedback in LLMs
• Large Language Models for Mathematical Reasoning: Progresses and Challenges
• SkVM: Revisiting Language VM for Skills across Heterogenous LLMs and Harnesses
• InternVLA-M1: A Spatially Guided Vision-Language-Action Framework for Generalist Robot Policy
• Mem0: Building Production-Ready AI Agents with Scalable Long-Term Memory
• AutoGuide: Automated Generation and Selection of Context-Aware Guidelines for Large Language Model Agents
• Synthetic Data Generation & Multi-Step RL for Reasoning & Tool Use
• Understanding the planning of LLM agents: A survey
• Search-o1: Agentic Search-Enhanced Large Reasoning Models
• A Comprehensive Survey on Reinforcement Learning-based Agentic Search: Foundations, Roles, Optimizations, Evaluations, and Applications
• SimpleMem: Efficient Lifelong Memory for LLM Agents
• SKILL0: In-Context Agentic Reinforcement Learning for Skill Internalization
• Large Language Model Agent: A Survey on Methodology, Applications and Challenges
• Skill-Pro: Learning Reusable Skills from Experience via Non-Parametric PPO for LLM Agents
• Trace2Skill: Distill Trajectory-Local Lessons into Transferable Agent Skills
• Proximal Policy Optimization Algorithms
• DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models
• LLM With Tools: A Survey
• Voyager: An Open-Ended Embodied Agent with Large Language Models
• WebXSkill: Skill Learning for Autonomous Web Agents

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