Disco-RAG: Discourse-Aware Retrieval-Augmented Generation
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2026
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866911599920414720 |
|---|---|
| author | Liu, Dongqi Ding, Hang Feng, Qiming Xie, Xurong Xue, Zhucun Wang, Chengjie Li, Jian Zhang, Jiangning Wang, Yabiao |
| author_facet | Liu, Dongqi Ding, Hang Feng, Qiming Xie, Xurong Xue, Zhucun Wang, Chengjie Li, Jian Zhang, Jiangning Wang, Yabiao |
| contents | Retrieval-Augmented Generation (RAG) has emerged as an important means of enhancing the performance of large language models (LLMs) in knowledge-intensive tasks. However, most existing RAG strategies treat retrieved passages in a flat and unstructured way, which prevents the model from capturing structural cues and constrains its ability to synthesize knowledge from dispersed evidence across documents. To overcome these limitations, we propose Disco-RAG, a discourse-aware framework that explicitly injects discourse signals into the generation process. Our method constructs intra-chunk discourse trees to capture local hierarchies and builds inter-chunk rhetorical graphs to model cross-passage coherence. These structures are jointly integrated into a planning blueprint that conditions the generation. Experiments on question answering and long-document summarization benchmarks show the efficacy of our approach. Disco-RAG achieves state-of-the-art results on the benchmarks without fine-tuning. These findings underscore the important role of discourse structure in advancing RAG systems. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2601_04377 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Disco-RAG: Discourse-Aware Retrieval-Augmented Generation Liu, Dongqi Ding, Hang Feng, Qiming Xie, Xurong Xue, Zhucun Wang, Chengjie Li, Jian Zhang, Jiangning Wang, Yabiao Computation and Language Artificial Intelligence Machine Learning Retrieval-Augmented Generation (RAG) has emerged as an important means of enhancing the performance of large language models (LLMs) in knowledge-intensive tasks. However, most existing RAG strategies treat retrieved passages in a flat and unstructured way, which prevents the model from capturing structural cues and constrains its ability to synthesize knowledge from dispersed evidence across documents. To overcome these limitations, we propose Disco-RAG, a discourse-aware framework that explicitly injects discourse signals into the generation process. Our method constructs intra-chunk discourse trees to capture local hierarchies and builds inter-chunk rhetorical graphs to model cross-passage coherence. These structures are jointly integrated into a planning blueprint that conditions the generation. Experiments on question answering and long-document summarization benchmarks show the efficacy of our approach. Disco-RAG achieves state-of-the-art results on the benchmarks without fine-tuning. These findings underscore the important role of discourse structure in advancing RAG systems. |
| title | Disco-RAG: Discourse-Aware Retrieval-Augmented Generation |
| topic | Computation and Language Artificial Intelligence Machine Learning |
| url | https://arxiv.org/abs/2601.04377 |