Symbolic-Neural Soft-Logic Reasoning: Towards Robust and Verifiable Thinking Chains via Cooperative Evolution
Rui Wang, Zeming Wei, Yihao Zhang, Xiaokun Luan · 2026
Read Symbolic-Neural Soft-Logic Reasoning: Towards Robust and Verifiable Thinking Chains via Cooperative Evolution by Rui Wang, Zeming Wei, Yihao Zhang, Xiaokun Luan (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-2605-25618.txt.
Description
Wang, Wei, Zhang and Luan tackle unfaithful chain-of-thought work, where step-by-step chains look sound without being reliable. They combine language models with symbolic logic but loosen strict rules, keeping results checkable while softening brittle either-or edges. Their SSR framework also builds verifiable, human-like logic chains for training. Tests across models and benchmarks show it beating current methods on robustness and interpretability.
Catalog record
| CREATOR | Rui Wang, Zeming Wei, Yihao Zhang, Xiaokun Luan |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 71827 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2605.25618