Less Finetuning, Better Retrieval: Rethinking LLM Adaptation for Biomedical Retrievers via Synthetic Data and Model Merging

• URL: http://arxiv.org/abs/2602.04731v1
• Date: Wed, 04 Feb 2026 16:36:00 GMT
• Title: Less Finetuning, Better Retrieval: Rethinking LLM Adaptation for Biomedical Retrievers via Synthetic Data and Model Merging
• Authors: Sameh Khattab, Jean-Philippe Corbeil, Osman Alperen Koraş, Amin Dada, Julian Friedrich, François Beaulieu, Paul Vozila, Jens Kleesiek,
• Abstract summary: Retrieval-augmented generation (RAG) has become the backbone of grounding Large Language Models (LLMs)<n>We present Synthesize-Merge-Train (STM), a modular framework that enhances decoder-only LLMs with synthetic hard negatives, retrieval prompt optimization, and model merging.<n> 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.
• Score: 6.761707146228743
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: 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.

Related papers

• Learn More, Forget Less: A Gradient-Aware Data Selection Approach for LLM [51.21051698747157]
  We propose a self-adaptive gradient-aware data selection approach (GrADS) for supervised fine-tuning of large language models (LLMs)<n>Specifically, we design self-guided criteria that leverage the magnitude and statistical distribution of gradients to prioritize examples that contribute the most to the model's learning process.<n>Through extensive experimentation with various LLMs across diverse domains such as medicine, law, and finance, GrADS has demonstrated significant efficiency and cost-effectiveness.
  arXiv  Detail & Related papers  (2025-11-07T08:34:50Z)
• Refining Hybrid Genetic Search for CVRP via Reinforcement Learning-Finetuned LLM [32.938753667649074]
  Large language models (LLMs) are increasingly used as automated designers for vehicle routing problems (VRPs)<n>This work challenges that paradigm by demonstrating that a smaller, specialized LLM, when meticulously fine-tuned, can generate components that surpass expert-crafteds within advanced solvers.<n>We propose RFTHGS, a novel Reinforcement learning framework for Fine-Tuning a small LLM to generate high-performance crossover operators.
  arXiv  Detail & Related papers  (2025-10-13T08:08:58Z)
• MM-HELIX: Boosting Multimodal Long-Chain Reflective Reasoning with Holistic Platform and Adaptive Hybrid Policy Optimization [103.74675519953898]
  Long-chain reflective reasoning is a prerequisite for solving complex real-world problems.<n>We build a benchmark consisting 1,260 samples of 42 challenging synthetic tasks.<n>We generate post-training data and explore learning paradigms for exploiting such data.
  arXiv  Detail & Related papers  (2025-10-09T17:53:58Z)
• Domain-Aware RAG: MoL-Enhanced RL for Efficient Training and Scalable Retrieval [5.640810636056805]
  MoLER is a domain-aware RAG method that uses MoL-Enhanced Reinforcement Learning to optimize retrieval.<n>MoLER bridges the knowledge gap in RAG systems, enabling robust and scalable retrieval in specialized domains.
  arXiv  Detail & Related papers  (2025-09-08T13:04:07Z)
• How Do LLM-Generated Texts Impact Term-Based Retrieval Models? [76.92519309816008]
  This paper investigates the influence of large language models (LLMs) on term-based retrieval models.<n>Our linguistic analysis reveals that LLM-generated texts exhibit smoother high-frequency and steeper low-frequency Zipf slopes.<n>Our study further explores whether term-based retrieval models demonstrate source bias, concluding that these models prioritize documents whose term distributions closely correspond to those of the queries.
  arXiv  Detail & Related papers  (2025-08-25T06:43:27Z)
• RAG in the Wild: On the (In)effectiveness of LLMs with Mixture-of-Knowledge Retrieval Augmentation [45.679455112940175]
  Retrieval-augmented generation (RAG) enhances large language models (LLMs) by integrating external knowledge retrieved at inference time.<n>We evaluated RAG systems using MassiveDS, a large-scale datastore with mixture of knowledge, and identified critical limitations.
  arXiv  Detail & Related papers  (2025-07-26T20:57:24Z)
• Mitigating Forgetting in LLM Fine-Tuning via Low-Perplexity Token Learning [65.23593936798662]
  We show that fine-tuning with LLM-generated data improves target task performance and reduces non-target task degradation.<n>This is the first work to provide an empirical explanation based on token perplexity reduction to mitigate catastrophic forgetting in LLMs after fine-tuning.
  arXiv  Detail & Related papers  (2025-01-24T08:18:56Z)
• SimRAG: Self-Improving Retrieval-Augmented Generation for Adapting Large Language Models to Specialized Domains [45.349645606978434]
  Retrieval-augmented generation (RAG) enhances the question-answering abilities of large language models (LLMs)<n>We propose SimRAG, a self-training approach that equips the LLM with joint capabilities of question answering and question generation for domain adaptation.<n> Experiments on 11 datasets, spanning two backbone sizes and three domains, demonstrate that SimRAG outperforms baselines by 1.2%--8.6%.
  arXiv  Detail & Related papers  (2024-10-23T15:24:16Z)
• Self-Exploring Language Models: Active Preference Elicitation for Online Alignment [88.56809269990625]
  We propose a bilevel objective optimistically biased towards potentially high-reward responses to actively explore out-of-distribution regions. Our experimental results demonstrate that when fine-tuned on Zephyr-7B-SFT and Llama-3-8B-Instruct models, Self-Exploring Language Models (SELM) significantly boosts the performance on instruction-following benchmarks.
  arXiv  Detail & Related papers  (2024-05-29T17:59:07Z)
• From Generalist to Specialist: Improving Large Language Models for Medical Physics Using ARCoT [0.0]
  ARCoT (Adaptable Retrieval-based Chain of Thought) is a framework designed to enhance the domain-specific accuracy of Large Language Models (LLMs) Our model outperformed standard LLMs and reported average human performance, demonstrating improvements of up to 68%.
  arXiv  Detail & Related papers  (2024-05-17T18:31:38Z)
• Omni-SMoLA: Boosting Generalist Multimodal Models with Soft Mixture of Low-rank Experts [74.40198929049959]
  Large multi-modal models (LMMs) exhibit remarkable performance across numerous tasks. generalist LMMs often suffer from performance degradation when tuned over a large collection of tasks. We propose Omni-SMoLA, an architecture that uses the Soft MoE approach to mix many multimodal low rank experts.
  arXiv  Detail & Related papers  (2023-12-01T23:04:27Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.