DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning

TL;DR

This paper demonstrates that reinforcement learning can incentivize reasoning in large language models, enabling them to develop advanced reasoning skills without human-labeled data, and improving performance on complex tasks.

Contribution

It introduces a reinforcement learning framework that promotes emergent reasoning abilities in LLMs without relying on human-annotated reasoning trajectories.

Findings

Models trained with RL outperform supervised counterparts on reasoning tasks.

Emergent reasoning patterns include self-reflection and verification.

Enhanced reasoning skills can be transferred to smaller models.

Abstract

General reasoning represents a long-standing and formidable challenge in artificial intelligence. Recent breakthroughs, exemplified by large language models (LLMs) and chain-of-thought prompting, have achieved considerable success on foundational reasoning tasks. However, this success is heavily contingent upon extensive human-annotated demonstrations, and models' capabilities are still insufficient for more complex problems. Here we show that the reasoning abilities of LLMs can be incentivized through pure reinforcement learning (RL), obviating the need for human-labeled reasoning trajectories. The proposed RL framework facilitates the emergent development of advanced reasoning patterns, such as self-reflection, verification, and dynamic strategy adaptation. Consequently, the trained model achieves superior performance on verifiable tasks such as mathematics, coding competitions, and STEM fields, surpassing its counterparts trained via conventional supervised learning on human demonstrations. Moreover, the emergent reasoning patterns exhibited by these large-scale models can be systematically harnessed to guide and enhance the reasoning capabilities of smaller models.