SC-Taxo: Hierarchical Taxonomy Generation under Semantic Consistency Constraints using Large Language Models
Shiqiang Cai, Nianhong Niu, Shizhu He, Kang Liu, Jun Zhao · 2026
Read SC-Taxo: Hierarchical Taxonomy Generation under Semantic Consistency Constraints using Large Language Models by Shiqiang Cai, Nianhong Niu, Shizhu He, Kang Liu, Jun Zhao (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2605-00620.txt.
Description
Shiqiang Cai and 4 co-authors build taxonomies for science, the layered maps of a research field, with large language models, keeping the levels honest in meaning. A good taxonomy speeds literature search and feeds trend tracking. Older generators, though, make structures whose levels drift in meaning, with headings that match neither parents nor siblings.
A study of real outputs blames weak tracking of whether levels mean the same thing. Their system, SC-Taxo, builds headings in both directions. It pulls general terms up from the leaves while pinning terms down from the roots, and models links among same-level headings to keep siblings in line. Cleanup stages that respect the levels then polish the result. On several test sets the method improved both how well levels line up and heading quality. Tests on Chinese science papers showed the approach carries across languages.
Catalog record
| CREATOR | Shiqiang Cai, Nianhong Niu, Shizhu He, Kang Liu, Jun Zhao |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 37 KB |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2605.00620