{"ID":5676017,"CreatedAt":"2026-07-03T01:40:09.565152011Z","UpdatedAt":"2026-07-04T22:11:50.569877204Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2607.01511","arxiv_id":"2607.01511","title":"Revisiting Chain-of-Thought Reasoning under Limited Supervision: Semi-supervised Chain-of-Thought Learning","abstract":"Chain-of-thought (CoT) reasoning has emerged as an effective approach for activating latent reasoning capabilities in large language models. However, most existing CoT methods use reasoning chains mainly as inference-time prompts, while the generated reasoning traces are rarely reused as semi-supervised learning signals. In this report, we define \\textbf{Semi-supervised Chain-of-Thought Learning} and propose \\textbf{Semi-CoT}, a simple framework that uses unlabeled questions to construct pseudo reasoning supervision. Semi-CoT samples multiple pseudo-CoTs for each unlabeled question, estimates answer-level semantic entropy, and selects low-entropy reasoning chains as reliable pseudo-CoT demonstrations. This extends the self-training view of CoT from inference-time refinement to semi-supervised pseudo-supervision. Pilot experiments on AQuA, SVAMP, GSM8K, and MultiArith show that the entropy gate selects high-precision pseudo-CoTs, with pseudo-answer precision ranging from $91.36\\%$ to $100\\%$. Semi-CoT also gives small gains on SVAMP and GSM8K, while AQuA shows negative transfer and MultiArith reaches a ceiling. These results suggest that unlabeled questions can provide reliable pseudo reasoning signals, but their effective use still requires stronger demonstration selection or student training.","short_abstract":"Chain-of-thought (CoT) reasoning has emerged as an effective approach for activating latent reasoning capabilities in large language models. However, most existing CoT methods use reasoning chains mainly as inference-time prompts, while the generated reasoning traces are rarely reused as semi-supervised learning signal...","url_abs":"https://arxiv.org/abs/2607.01511","url_pdf":"https://arxiv.org/pdf/2607.01511v1","authors":"[\"Hongyang He\",\"Jiuming Liu\",\"Victor Sanchez\"]","published":"2026-07-01T22:17:39Z","proceeding":"cs.AI","tasks":"[\"cs.AI\",\"cs.LG\"]","methods":"[\"Language Model\"]","has_code":false}