Salvato in:
Dettagli Bibliografici
Autori principali: Zhang, Jie, Tang, Bo, Shao, Wanzi, Wei, Wenqiang, Zhao, Jihao, Zhu, Jianqing, li, Zhiyu, Xi, Wen, Lin, Zehao, Xiong, Feiyu, Tan, Yanchao
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:https://arxiv.org/abs/2511.12520
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866912712629420032
author Zhang, Jie
Tang, Bo
Shao, Wanzi
Wei, Wenqiang
Zhao, Jihao
Zhu, Jianqing
li, Zhiyu
Xi, Wen
Lin, Zehao
Xiong, Feiyu
Tan, Yanchao
author_facet Zhang, Jie
Tang, Bo
Shao, Wanzi
Wei, Wenqiang
Zhao, Jihao
Zhu, Jianqing
li, Zhiyu
Xi, Wen
Lin, Zehao
Xiong, Feiyu
Tan, Yanchao
contents Retrieval-Augmented Generation (RAG) improves large language models by retrieving external knowledge, often truncated into smaller chunks due to the input context window, which leads to information loss, resulting in response hallucinations and broken reasoning chains. Moreover, traditional RAG retrieves unstructured knowledge, introducing irrelevant details that hinder accurate reasoning. To address these issues, we propose TAdaRAG, a novel RAG framework for on-the-fly task-adaptive knowledge graph construction from external sources. Specifically, we design an intent-driven routing mechanism to a domain-specific extraction template, followed by supervised fine-tuning and a reinforcement learning-based implicit extraction mechanism, ensuring concise, coherent, and non-redundant knowledge integration. Evaluations on six public benchmarks and a real-world business benchmark (NowNewsQA) across three backbone models demonstrate that TAdaRAG outperforms existing methods across diverse domains and long-text tasks, highlighting its strong generalization and practical effectiveness.
format Preprint
id arxiv_https___arxiv_org_abs_2511_12520
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle TAdaRAG: Task Adaptive Retrieval-Augmented Generation via On-the-Fly Knowledge Graph Construction
Zhang, Jie
Tang, Bo
Shao, Wanzi
Wei, Wenqiang
Zhao, Jihao
Zhu, Jianqing
li, Zhiyu
Xi, Wen
Lin, Zehao
Xiong, Feiyu
Tan, Yanchao
Computation and Language
Retrieval-Augmented Generation (RAG) improves large language models by retrieving external knowledge, often truncated into smaller chunks due to the input context window, which leads to information loss, resulting in response hallucinations and broken reasoning chains. Moreover, traditional RAG retrieves unstructured knowledge, introducing irrelevant details that hinder accurate reasoning. To address these issues, we propose TAdaRAG, a novel RAG framework for on-the-fly task-adaptive knowledge graph construction from external sources. Specifically, we design an intent-driven routing mechanism to a domain-specific extraction template, followed by supervised fine-tuning and a reinforcement learning-based implicit extraction mechanism, ensuring concise, coherent, and non-redundant knowledge integration. Evaluations on six public benchmarks and a real-world business benchmark (NowNewsQA) across three backbone models demonstrate that TAdaRAG outperforms existing methods across diverse domains and long-text tasks, highlighting its strong generalization and practical effectiveness.
title TAdaRAG: Task Adaptive Retrieval-Augmented Generation via On-the-Fly Knowledge Graph Construction
topic Computation and Language
url https://arxiv.org/abs/2511.12520