FactNet: A Billion-Scale Knowledge Graph for Multilingual Factual Grounding
Yingli Shen, Wen Lai, Jie Zhou, Xueren Zhang, Yudong Wang, Kangyang Luo, Shuo Wang, Ge Gao, Alexander Fraser, Maosong Sun · 2026
Read FactNet: A Billion-Scale Knowledge Graph for Multilingual Factual Grounding by Yingli Shen, Wen Lai, Jie Zhou, Xueren Zhang, Yudong Wang, Kangyang Luo, Shuo Wang, Ge Gao, Alexander Fraser, Maosong Sun (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-2602-03417.txt.
Description
Yingli Shen and 9 co-authors introduce FactNet, a knowledge resource meant to keep language models factually grounded. Their 2026 paper describes a graph that links 1.7 billion Wikidata assertions to just over 3 billion evidence pointers. The evidence comes from 316 native-language editions of Wikipedia, far beyond the usual English-only resources.
A deterministic pipeline builds the graph, so every evidence unit traces back to its exact source, down to the byte. The team also releases FactNet-Bench, a test suite for completing knowledge graphs, answering questions, and checking facts, with controls against data leakage. Experiments show the bench separates method families clearly, and that cross-lingual structure helps knowledge move between languages.
Catalog record
| CREATOR | Yingli Shen, Wen Lai, Jie Zhou, Xueren Zhang, Yudong Wang, Kangyang Luo, Shuo Wang, Ge Gao, Alexander Fraser, Maosong Sun |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 122 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2602-03417