Fine-tuning with RAG for Improving LLM Learning of New Skills
This addresses the computational overhead and maintenance requirements of retrieval-augmented generation for LLM agents in interactive tasks, though it appears incremental as it builds on existing distillation and RAG techniques.
The paper tackles the problem of LLM agents failing in predictable ways during multi-step tasks by proposing a pipeline that converts inference-time retrieval into learned competence through distillation, achieving up to 91% success on ALFWorld (vs. 79% for baselines) and improving WebShop scores to 72 (vs. 61 for baselines) while using 10-60% fewer tokens.
Large language model (LLM) agents deployed for multi-step tasks frequently fail in predictable ways: attempting actions with unmet preconditions, issuing redundant commands, or mishandling environment constraints. While retrieval-augmented generation (RAG) can improve performance by providing runtime guidance, it requires maintaining external knowledge databases and adds computational overhead at every deployment. We propose a simple pipeline that converts inference-time retrieval into learned competence through distillation. Our approach: (1) extracts compact, reusable hints from agent failures, (2) uses these hints to generate improved teacher trajectories via one-shot retrieval at episode start, and (3) trains student models on these trajectories with hint strings removed, forcing internalization rather than memorization. Across two interactive benchmarks, ALFWorld (household tasks) and WebShop (online shopping), distilled students consistently outperform baseline agents, achieving up to 91% success on ALFWorld (vs. 79% for baselines) and improving WebShop scores to 72 (vs. 61 for baselines), while using 10-60% fewer tokens than retrieval-augmented teachers depending on the environment. The approach generalizes across model scales (7B/14B parameters) and agent architectures (ReAct/StateAct), demonstrating that retrieval benefits can be effectively internalized through targeted fine-tuning without permanent runtime dependencies.