Revisiting the Capacity Gap in Chain-of-Thought Distillation from a Practical Perspective

Tokio Kajitsuka, Ukyo Honda, Sho Takase


Abstract
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.
Anthology ID:
2026.emnlp-main.2265
Volume:
Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing
Month:
October
Year:
2026
Address:
Budapest, Hungary
Editors:
Hung-Yi Lee, Sunipa Dev, Jessy Li, Desmond Elliott
Venue:
EMNLP
SIG:
SIGDAT
Publisher:
Association for Computational Linguistics
Note:
Pages:
51505–51528
Language:
URL:
https://aclanthology.org/2026.emnlp-main.2265/
DOI:
Bibkey:
Cite (ACL):
Tokio Kajitsuka, Ukyo Honda, and Sho Takase. 2026. Revisiting the Capacity Gap in Chain-of-Thought Distillation from a Practical Perspective. In Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing, pages 51505–51528, Budapest, Hungary. Association for Computational Linguistics.
Cite (Informal):
Revisiting the Capacity Gap in Chain-of-Thought Distillation from a Practical Perspective (Kajitsuka et al., EMNLP 2026)
Copy Citation:
PDF:
https://aclanthology.org/2026.emnlp-main.2265.pdf