Reasoning with OmniThought: A Large CoT Dataset with Verbosity and Cognitive Difficulty Annotations

TL;DR

OmniThought is a large-scale dataset with 2 million chain-of-thought reasoning processes, annotated with verbosity and difficulty scores, designed to improve large reasoning models' training and reasoning abilities.

Contribution

The paper introduces OmniThought, a comprehensive CoT dataset with novel annotations, and demonstrates its effectiveness in enhancing large reasoning models.

Findings

Scores improve LRM training effectiveness

Enhanced reasoning abilities in trained LRMs

Optimal CoT verbosity and difficulty levels identified

Abstract

The emergence of large reasoning models (LRMs) has transformed Natural Language Processing by excelling in complex tasks such as mathematical problem-solving and code generation. These models leverage chain-of-thought (CoT) processes, enabling them to emulate human-like reasoning strategies. However, the advancement of LRMs is hindered by the lack of comprehensive CoT datasets. Current resources often fail to provide extensive reasoning problems with coherent CoT processes distilled from multiple teacher models and do not account for multifaceted properties describing the internal characteristics of CoTs. To address these challenges, we introduce OmniThought, a large-scale dataset featuring 2 million CoT processes generated and validated by two powerful LRMs as teacher models. Each CoT process in OmniThought is annotated with novel Reasoning Verbosity (RV) and Cognitive Difficulty (CD) scores, which describe the appropriateness of CoT verbosity and cognitive difficulty level for models to comprehend these reasoning processes. We further establish a self-reliant pipeline to curate this dataset. Extensive experiments using Qwen2.5 models of various sizes demonstrate the positive impact of our proposed scores on LRM training effectiveness. Based on the proposed OmniThought dataset, we further train and release a series of high-performing LRMs, specifically equipped with stronger reasoning abilities and optimal CoT output length and difficulty level. Our contributions significantly enhance the development and training of LRMs for solving complex tasks.