paper-with-me

홈 › Papers

Enhancing LLMs' Reasoning-Intensive Multimedia Search Capabilities through Fine-Tuning and Reinforcement Learning

2025-05-24 · Jinzheng Li, Sibo Ju, Yanzhou Su, Hongguang Li, Yiqing Shen

Existing large language models (LLMs) driven search agents typically rely on prompt engineering to decouple the user queries into search plans, limiting their effectiveness in complex scenarios requiring reasoning. Furthermore, they suffer from excessive token consumption due to Python-based search plan representations and inadequate integration of multimedia elements for both input processing and response generation. To address these challenges, we introduce SearchExpert, a training method for LLMs to improve their multimedia search capabilities in response to complex search queries. Firstly, we reformulate the search plan in an efficient natural language representation to reduce token consumption. Then, we propose the supervised fine-tuning for searching (SFTS) to fine-tune LLM to adapt to these representations, together with an automated dataset construction pipeline. Secondly, to improve reasoning-intensive search capabilities, we propose the reinforcement learning from search feedback (RLSF) that takes the search results planned by LLM as the reward signals. Thirdly, we propose a multimedia understanding and generation agent that enables the fine-tuned LLM to process visual input and produce visual output during inference. Finally, we establish an automated benchmark construction pipeline and a human evaluation framework. Our resulting benchmark, SearchExpertBench-25, comprises 200 multiple-choice questions spanning financial and international news scenarios that require reasoning in searching. Experiments demonstrate that SearchExpert outperforms the commercial LLM search method (Perplexity Pro) by 36.60% on the existing FinSearchBench-24 benchmark and 54.54% on our proposed SearchExpertBench-25. Human evaluations further confirm the superior readability.

📄 PDF Abstract BibTeX arXiv:2505.18831

Code (0)

등록된 구현이 없습니다.

Tasks

Multiple-choicePrompt EngineeringResponse Generation

Similar Papers 제목 키워드 기반

Enhancing LLMs' Clinical Reasoning with Real-World Data from a Nationwide Sepsis Registry

2025-05-05 · Junu Kim, Chaeeun Shim, Sungjin Park, Su Yeon Lee 외

Although large language models (LLMs) have demonstrated impressive reasoning capabilities across general domains, their effectiveness in real-world clinical practice remains limited. This is likely due to their insuffici…

StructRAG: Boosting Knowledge Intensive Reasoning of LLMs via Inference-time Hybrid Information Structurization

2024-10-11 · Zhuoqun Li, Xuanang Chen, Haiyang Yu, Hongyu Lin 외

Retrieval-augmented generation (RAG) is a key means to effectively enhance large language models (LLMs) in many knowledge-based tasks. However, existing RAG methods struggle with knowledge-intensive reasoning tasks, beca…

RAGRetrieval-augmented Generation

GeoPQA: Bridging the Visual Perception Gap in MLLMs for Geometric Reasoning

2025-09-22 · Guizhen Chen, Weiwen Xu, Hao Zhang, Hou Pong Chan 외 arxiv

Recent advancements in reinforcement learning (RL) have enhanced the reasoning abilities of large language models (LLMs), yet the impact on multimodal LLMs (MLLMs) is limited. Particularly in vision-intensive tasks like …

Reinforcement Learning

MCTS-RAG: Enhancing Retrieval-Augmented Generation with Monte Carlo Tree Search

2025-03-26 · Yunhai Hu, Yilun Zhao, Chen Zhao, Arman Cohan

We introduce MCTS-RAG, a novel approach that enhances the reasoning capabilities of small language models on knowledge-intensive tasks by leveraging retrieval-augmented generation (RAG) to provide relevant context and Mo…

Decision MakingRAGRetrievalRetrieval-augmented Generation

RARE: Retrieval-Augmented Reasoning Enhancement for Large Language Models

2024-12-03 · Hieu Tran, Zonghai Yao, Junda Wang, Yifan Zhang 외

This work introduces RARE (Retrieval-Augmented Reasoning Enhancement), a versatile extension to the mutual reasoning framework (rStar), aimed at enhancing reasoning accuracy and factual integrity across large language mo…

Information RetrievalRetrieval