Reinforcement learning· 2024
DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models
On the reading lists
- #306 in the Top 500 (2026 edition)
- #15 in Reinforcement learning (2026 edition)
Why it matters
Introduced group relative policy optimization, simplifying RL training for mathematical reasoning.
Condensed from the expert lists' notes.
The record
- First released
- 2024 (from arXiv 2402.03300)
- Citations
- 10,104 citations, as counted by Papers With Code (CC BY-SA 4.0)
- Named by
- the chatgpt list only
- Also relevant to
- Language, speech & LLMs
- Score
- 0.440 · computed rank 306 · consensus 0.3, must-read 0, citations percentile 1 (2020s · reinforcement-learning), awards 0, existing lists 0. Method
Data sources
Catalogue data from OpenAlex (CC0), Crossref (CC0 metadata; no abstracts are shown), arXiv, and Papers With Code (CC BY-SA 4.0). Each count names its source. See the method.