Researchers propose EvoTool, a gradient-free evolutionary framework that optimizes tool-use policies in LLM agents by decomposing them into four modules and iteratively improving each through blame attribution and targeted mutation. The approach outperforms GPT-4.1 and Qwen3-8B baselines by over 5 percentage points across four benchmarks.
Researchers have proposed MAGE, a meta-reinforcement learning framework that enables large language model agents to adapt and strategize in dynamic environments. Unlike existing approaches that struggle with long-term adaptation, MAGE embeds the learning process directly within the model by integrating interaction histories and reflections into the context window.
Researchers introduce RAPO (Retrieval-Augmented Policy Optimization), a reinforcement learning framework that improves LLM agent reasoning by incorporating off-policy retrieval signals during training. The method achieves an average 5.0% performance gain across fourteen datasets and delivers 1.2x faster training efficiency compared to existing agentic RL approaches.
DeepXiv-SDK introduces a three-layer agentic data interface designed to give LLM agents efficient, cost-aware access to scientific literature. The system transforms unstructured data into normalized JSON, offers retrieval tools via CLI, MCP, and Python SDK, and currently covers the complete arXiv corpus with daily synchronization.