Salvato in:
| Autori principali: | , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | https://arxiv.org/abs/2509.25814 |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866914067387514880 |
|---|---|
| author | Kim, Boyoung Lee, Dosung An, Sumin Jeong, Jinseong Seo, Paul Hongsuck |
| author_facet | Kim, Boyoung Lee, Dosung An, Sumin Jeong, Jinseong Seo, Paul Hongsuck |
| contents | Recent advances in question answering have led to substantial progress in tasks such as multi-hop reasoning. However, global sensemaking-answering questions by synthesizing information from an entire corpus remains a significant challenge. A prior graph-based approach to global sensemaking lacks retrieval mechanisms, topic specificity, and incurs high inference costs. To address these limitations, we propose ReTAG, a Retrieval-Enhanced, Topic-Augmented Graph framework that constructs topic-specific subgraphs and retrieves the relevant summaries for response generation. Experiments show that ReTAG improves response quality while significantly reducing inference time compared to the baseline. Our code is available at https://github.com/bykimby/retag. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2509_25814 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | ReTAG: Retrieval-Enhanced, Topic-Augmented Graph-Based Global Sensemaking Kim, Boyoung Lee, Dosung An, Sumin Jeong, Jinseong Seo, Paul Hongsuck Computation and Language Recent advances in question answering have led to substantial progress in tasks such as multi-hop reasoning. However, global sensemaking-answering questions by synthesizing information from an entire corpus remains a significant challenge. A prior graph-based approach to global sensemaking lacks retrieval mechanisms, topic specificity, and incurs high inference costs. To address these limitations, we propose ReTAG, a Retrieval-Enhanced, Topic-Augmented Graph framework that constructs topic-specific subgraphs and retrieves the relevant summaries for response generation. Experiments show that ReTAG improves response quality while significantly reducing inference time compared to the baseline. Our code is available at https://github.com/bykimby/retag. |
| title | ReTAG: Retrieval-Enhanced, Topic-Augmented Graph-Based Global Sensemaking |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2509.25814 |