Retrieval Feedback Memory Enhancement Large Model Retrieval Generation Method
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866916915804372992 |
|---|---|
| author | Li, Leqian Shi, Dianxi Zhou, Jialu Wei, Xinyu Yang, Mingyue Jin, Songchang Yang, Shaowu |
| author_facet | Li, Leqian Shi, Dianxi Zhou, Jialu Wei, Xinyu Yang, Mingyue Jin, Songchang Yang, Shaowu |
| contents | Large Language Models (LLMs) have shown remarkable capabilities across diverse tasks, yet they face inherent limitations such as constrained parametric knowledge and high retraining costs. Retrieval-Augmented Generation (RAG) augments the generation process by retrieving externally stored knowledge absent from the models internal parameters. However, RAG methods face challenges such as information loss and redundant retrievals during multi-round queries, accompanying the difficulties in precisely characterizing knowledge gaps for complex tasks. To address these problems, we propose Retrieval Feedback and Memory Retrieval Augmented Generation(RFM-RAG), which transforms the stateless retrieval of previous methods into stateful continuous knowledge management by constructing a dynamic evidence pool. Specifically, our method generates refined queries describing the models knowledge gaps using relational triples from questions and evidence from the dynamic evidence pool; Retrieves critical external knowledge to iteratively update this evidence pool; Employs a R-Feedback Model to evaluate evidence completeness until convergence. Compared to traditional RAG methods, our approach enables persistent storage of retrieved passages and effectively distills key information from passages to construct clearly new queries. Experiments on three public QA benchmarks demonstrate that RFM-RAG outperforms previous methods and improves overall system accuracy. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2508_17862 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Retrieval Feedback Memory Enhancement Large Model Retrieval Generation Method Li, Leqian Shi, Dianxi Zhou, Jialu Wei, Xinyu Yang, Mingyue Jin, Songchang Yang, Shaowu Information Retrieval Large Language Models (LLMs) have shown remarkable capabilities across diverse tasks, yet they face inherent limitations such as constrained parametric knowledge and high retraining costs. Retrieval-Augmented Generation (RAG) augments the generation process by retrieving externally stored knowledge absent from the models internal parameters. However, RAG methods face challenges such as information loss and redundant retrievals during multi-round queries, accompanying the difficulties in precisely characterizing knowledge gaps for complex tasks. To address these problems, we propose Retrieval Feedback and Memory Retrieval Augmented Generation(RFM-RAG), which transforms the stateless retrieval of previous methods into stateful continuous knowledge management by constructing a dynamic evidence pool. Specifically, our method generates refined queries describing the models knowledge gaps using relational triples from questions and evidence from the dynamic evidence pool; Retrieves critical external knowledge to iteratively update this evidence pool; Employs a R-Feedback Model to evaluate evidence completeness until convergence. Compared to traditional RAG methods, our approach enables persistent storage of retrieved passages and effectively distills key information from passages to construct clearly new queries. Experiments on three public QA benchmarks demonstrate that RFM-RAG outperforms previous methods and improves overall system accuracy. |
| title | Retrieval Feedback Memory Enhancement Large Model Retrieval Generation Method |
| topic | Information Retrieval |
| url | https://arxiv.org/abs/2508.17862 |