DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning research paper by DeepSeek, 2025

DeepSeek · Jan 22, 2025 · Reasoning · 5,691 citations · 462 upvotes

Reinforcement learning taught a model long step-by-step reasoning on par with OpenAI o1.

Summarised by hand from the abstract.

More from DeepSeek

Date	Paper	Topic	Citations
Sep 17, 2026	DeepSeek-V4.1-Flash: Pushing the Limits of KV Cache CompressionA 552B mixture-of-experts model with a 1M-token context, built to shrink the KV cache for agent workloads.Inference and efficiency · Sep 2026	Inference and efficiency	7

Topic

Date	Paper	Lab	Citations
Jan 28, 2022	Chain-of-Thought Prompting Elicits Reasoning in Large Language ModelsAsking a model to write out its intermediate steps (chain of thought) sharply improves its math and logic answers.Google · Jan 2022Google	Google	22.1k

About this paper

Authors: DeepSeek-AI, Daya Guo, Dejian Yang and 197 more
arXiv: 2501.12948 · PDF
Venue: Nature
Citations: 5,691, 921 influential · Semantic Scholar
Upvotes: 462 · Hugging Face
Code: github.com/deepseek-ai/deepseek-r1
Lab: DeepSeek

Changes

Date	What changed
Sep 24, 2026	New paperAdded to the listSep 24, 2026

Sources: each lab's own papers and arXiv, with citation and upvote counts from Semantic Scholar and Hugging Face. One-line summaries are for orientation, not a substitute for the paper. Logos via logo.dev; trademarks belong to their owners.

What it shows

More from DeepSeek

Other reasoning papers

Changes

New papers by email