Effectiveness of Chain-of-Thought in Distilling Reasoning Capability from Large Language Models

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Do, Cong-Thanh, Doddipatla, Rama, Knill, Kate
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866918190579187712
author Do, Cong-Thanh
Doddipatla, Rama
Knill, Kate
author_facet Do, Cong-Thanh
Doddipatla, Rama
Knill, Kate
contents Chain-of-Thought (CoT) prompting is a widely used method to improve the reasoning capability of Large Language Models (LLMs). More recently, CoT has been leveraged in Knowledge Distillation (KD) to transfer reasoning capability from a larger LLM to a smaller one. This paper examines the role of CoT in distilling the reasoning capability from larger LLMs to smaller LLMs using white-box KD, analysing its effectiveness in improving the performance of the distilled models for various natural language reasoning and understanding tasks. We conduct white-box KD experiments using LLMs from the Qwen and Llama2 families, employing CoT data from the CoT-Collection dataset. The distilled models are then evaluated on natural language reasoning and understanding tasks from the BIG-Bench-Hard (BBH) benchmark, which presents complex challenges for smaller LLMs. Experimental results demonstrate the role of CoT in improving white-box KD effectiveness, enabling the distilled models to achieve better average performance in natural language reasoning and understanding tasks from BBH.
format Preprint
id arxiv_https___arxiv_org_abs_2511_05184
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Effectiveness of Chain-of-Thought in Distilling Reasoning Capability from Large Language Models
Do, Cong-Thanh
Doddipatla, Rama
Knill, Kate
Computation and Language
Chain-of-Thought (CoT) prompting is a widely used method to improve the reasoning capability of Large Language Models (LLMs). More recently, CoT has been leveraged in Knowledge Distillation (KD) to transfer reasoning capability from a larger LLM to a smaller one. This paper examines the role of CoT in distilling the reasoning capability from larger LLMs to smaller LLMs using white-box KD, analysing its effectiveness in improving the performance of the distilled models for various natural language reasoning and understanding tasks. We conduct white-box KD experiments using LLMs from the Qwen and Llama2 families, employing CoT data from the CoT-Collection dataset. The distilled models are then evaluated on natural language reasoning and understanding tasks from the BIG-Bench-Hard (BBH) benchmark, which presents complex challenges for smaller LLMs. Experimental results demonstrate the role of CoT in improving white-box KD effectiveness, enabling the distilled models to achieve better average performance in natural language reasoning and understanding tasks from BBH.
title Effectiveness of Chain-of-Thought in Distilling Reasoning Capability from Large Language Models
topic Computation and Language
url https://arxiv.org/abs/2511.05184