Saved in:
| Main Authors: | , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2510.21900 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866914114105769984 |
|---|---|
| author | Zhang, Hongbo Cui, Han Wang, Yidong Tian, Yijian Guo, Qi Wang, Cunxiang Wu, Jian Song, Chiyu Zhang, Yue |
| author_facet | Zhang, Hongbo Cui, Han Wang, Yidong Tian, Yijian Guo, Qi Wang, Cunxiang Wu, Jian Song, Chiyu Zhang, Yue |
| contents | Automatic literature survey generation has attracted increasing attention, yet most existing systems follow a one-shot paradigm, where a large set of papers is retrieved at once and a static outline is generated before drafting. This design often leads to noisy retrieval, fragmented structures, and context overload, ultimately limiting survey quality. Inspired by the iterative reading process of human researchers, we propose \ours, a framework based on recurrent outline generation, in which a planning agent incrementally retrieves, reads, and updates the outline to ensure both exploration and coherence. To provide faithful paper-level grounding, we design paper cards that distill each paper into its contributions, methods, and findings, and introduce a review-and-refine loop with visualization enhancement to improve textual flow and integrate multimodal elements such as figures and tables. Experiments on both established and emerging topics show that \ours\ substantially outperforms state-of-the-art baselines in content coverage, structural coherence, and citation quality, while producing more accessible and better-organized surveys. To provide a more reliable assessment of such improvements, we further introduce Survey-Arena, a pairwise benchmark that complements absolute scoring and more clearly positions machine-generated surveys relative to human-written ones. The code is available at https://github.com/HancCui/IterSurvey\_Autosurveyv2. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2510_21900 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Deep Literature Survey Automation with an Iterative Workflow Zhang, Hongbo Cui, Han Wang, Yidong Tian, Yijian Guo, Qi Wang, Cunxiang Wu, Jian Song, Chiyu Zhang, Yue Computation and Language Artificial Intelligence Automatic literature survey generation has attracted increasing attention, yet most existing systems follow a one-shot paradigm, where a large set of papers is retrieved at once and a static outline is generated before drafting. This design often leads to noisy retrieval, fragmented structures, and context overload, ultimately limiting survey quality. Inspired by the iterative reading process of human researchers, we propose \ours, a framework based on recurrent outline generation, in which a planning agent incrementally retrieves, reads, and updates the outline to ensure both exploration and coherence. To provide faithful paper-level grounding, we design paper cards that distill each paper into its contributions, methods, and findings, and introduce a review-and-refine loop with visualization enhancement to improve textual flow and integrate multimodal elements such as figures and tables. Experiments on both established and emerging topics show that \ours\ substantially outperforms state-of-the-art baselines in content coverage, structural coherence, and citation quality, while producing more accessible and better-organized surveys. To provide a more reliable assessment of such improvements, we further introduce Survey-Arena, a pairwise benchmark that complements absolute scoring and more clearly positions machine-generated surveys relative to human-written ones. The code is available at https://github.com/HancCui/IterSurvey\_Autosurveyv2. |
| title | Deep Literature Survey Automation with an Iterative Workflow |
| topic | Computation and Language Artificial Intelligence |
| url | https://arxiv.org/abs/2510.21900 |