Revisiting the Capacity Gap in Chain-of-Thought Distillation from a Practical Perspective
On the map: LLM reasoning → Prompting and chain-of-thought reasoning
also in 1 more
Abstract (arXiv)
Chain-of-thought (CoT) distillation transfers reasoning behaviors from a strong teacher to a smaller student, but prior work reports a capacity gap: distillation may fail when the teacher-student capability mismatch is large. We revisit the capacity gap from a practical perspective by re-examining commonly used experimental settings. Notably, we find that CoT distillation often degrades performance compared to the student's pre-distillation baseline, and that some settings used in prior work, while suitable for establishing the capacity gap as a phenomenon, do not reflect realistic deployment scenarios. Complementing prior work that establishes the capacity gap, we evaluate its practical impact under more realistic settings and find that it does not consistently dominate; stronger teachers tend to be preferable when candidate teachers differ substantially in performance. Our results offer practical guidance for selecting teacher-student pairs in CoT distillation.
Key facts. Cited by 2 papers in the map (data through 14 Oct 2026). Placed in Prompting and chain-of-thought reasoning, one of 17 directions in LLM reasoning. Builds on 6 of its 38 references that are in the map. First posted on arXiv 10 Apr 2026.
Builds on
6 of 38 references.
References · 6
- 2015Distilling the Knowledge in a Neural Networkcited by 10,324
- 2021Training Verifiers to Solve Math Word Problemscited by 10,109
- 2024Qwen2.5 Technical Reportcited by 4,942
- 2024Gemma 2: Improving Open Language Models at a Practical Sizecited by 1,998
- 2023Orca: Progressive Learning from Complex Explanation Traces of GPT-4cited by 379
- 2015Unifying distillation and privileged informationcited by 233
Built on by
Citing papers per year: 2026: 2.
2 papers.
Similar, not cited
Close in wording, no citation either way
- 2025Unveiling the Key Factors for Distilling Chain-of-Thought Reasoning
- 2023Symbolic Chain-of-Thought Distillation: Small Models Can Also "Think" Step-by-Step
- 2026Improving Reasoning Capabilities in Small Models through Mixture-of-Layers Distillation with Stepwise Attention on Key Information
- 2026Knowledge Distillation Must Account for What It Loses
- 2025Merge-of-Thought Distillation
- 2026When Compression Helps and When It Hurts: Condition-Aware Analysis of Chain-of-Thought Distillation
- 2026Solving Without Stopping: On-Policy Distillation at Small Scale
- 2024Investigating Mysteries of CoT-Augmented Distillation