Less Finetuning, Better Retrieval: Rethinking LLM Adaptation for Biomedical Retrievers via Synthetic Data and Model Merging
This work addresses the challenge of efficiently specializing LLMs for retrieval in specialized domains like biomedicine, offering a scalable solution that is incremental in nature.
The paper tackles the problem of adapting general-purpose LLMs into effective domain-specific retrievers for biomedical applications, presenting the Synthesize-Train-Merge (STM) framework that boosts task-specific experts by up to 23.5% (average 7.5%) on medical and general tasks.
Retrieval-augmented generation (RAG) has become the backbone of grounding Large Language Models (LLMs), improving knowledge updates and reducing hallucinations. Recently, LLM-based retriever models have shown state-of-the-art performance for RAG applications. However, several technical aspects remain underexplored on how to adapt general-purpose LLMs into effective domain-specific retrievers, especially in specialized domains such as biomedicine. We present Synthesize-Train-Merge (STM), a modular framework that enhances decoder-only LLMs with synthetic hard negatives, retrieval prompt optimization, and model merging. Experiments on a subset of 12 medical and general tasks from the MTEB benchmark show STM boosts task-specific experts by up to 23.5\% (average 7.5\%) and produces merged models that outperform both single experts and strong baselines without extensive pretraining. Our results demonstrate a scalable, efficient path for turning general LLMs into high-performing, domain-specialized retrievers, preserving general-domain capabilities while excelling on specialized tasks.