arXiv：cs.AI· Petr Senatorov, Karim Vafin, Elena Tutubalina, Oleg Somov·· 10 小时前
Structural Chain-of-Thought：用可执行计算图替代自由形式推理
Structural Chain-of-Thought: Reasoning with Executable Computational Graphs
AI 导读
研究者提出 Structural Chain-of-Thought 框架，让模型生成可执行计算图而非自由形式推理，最终答案由确定性地执行该图得出，并采用基于图的强化学习训练，奖励结构有效性、答案正确性和与参考图的节点重叠。
正文
Abstract:Chain-of-thought prompting enables reasoning in large language models but produces explanations that are difficult to verify, expensive to generate, and not always faithful. We introduce a Structural Chain-of-Thought framework where models generate executable computational graphs instead of free-form reasoning. The final answer is obtained by deterministically executing the graph. The model is trained using graph-based reinforcement learning that rewards structural validity, answer correctness, and node overlap with reference graphs.
Experiments across five benchmarks and four model families show that our method consistently improves accuracy over supervised graph training and can reduce generation length by up to 5-6x compared with standard GRPO and chain-of-thought supervised training. Our faithfulness experiments reveal that models trained with our approach tend to rely on deriving their answers from the generated graphs rather than producing post-hoc rationalizations as usually in chain-of-thought reasoning.
| Subjects: | Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2610.15107 [cs.AI] |
| (or arXiv:2610.15107v1 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2610.15107 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Peter Senatorov [view email]
[v1]
Sat, 10 Oct 2026 18:29:30 UTC (3,070 KB)
来源：arXiv：cs.AI · arxiv.org