arxiv Moonshot AI (arXiv) · 916d ago

DeepSeekMath 7B achieves 51.7% on MATH benchmark using GRPO and curated data

from English

The DeepSeekMath paper introduces a 7B parameter open language model designed to enhance mathematical reasoning capabilities. The model is created by continuing the pre-training of DeepSeek-Coder-Base-v1.5 with 120B math-related tokens from Common Crawl, alongside natural language and code data.

DeepSeekMath 7B scores 51.7% on the competition-level MATH benchmark without external toolkits or voting techniques.
Self-consistency over 64 samples reaches 60.9% on the MATH benchmark.
The performance is attributed to a curated data selection pipeline and Group Relative Policy Optimization (GRPO).
GRPO is a variant of Proximal Policy Optimization that improves reasoning while optimizing memory usage.

The authors consider this significant because the model approaches the performance level of proprietary models like Gemini-Ultra and GPT-4.

Importance 1/3 Moonshot AI (arXiv) DeepSeek Reasoning models Research paper

Benchmarks

Benchmark	Model	Score
MATH-500	DeepSeekMath 7B	51.7%

Read original

arxiv Moonshot AI (arXiv) · 555d ago

Simple Scaling Lab releases s1-32B with test-time scaling via budget forcing

The Simple Scaling lab introduces s1-32B, a reasoning model that achieves strong performance through supervised fine-tuning on a curated dataset of 1,000 samples and a technique called budget forcing. This approach controls test-time compute by either terminating the model's thinking process or appending "Wait" to extend it, allowing the model to double-check its answers.

media Don't Worry About the Vase · 5d ago · 5 views

OpenAI's Astra solves 10 major open mathematics problems

OpenAI has released results from its unreleased internal model, Astra, which successfully solved ten significant open problems in mathematics. The solutions were generated by the model and subsequently formalized into Lean certificates by human researchers.

blog Simon Willison · 7d ago · 12 views

OpenAI's Astra model solves ten long-standing math problems using GPT-5.6 Sol

OpenAI has released solutions to ten mathematical and theoretical computer science problems that had seen no progress on their main results for at least a decade, using an internal version of its next major model, Astra.

arxiv arXiv cs.CL · 12d ago · 2 views

PIVOT shares prefix scans across query groups to accelerate token-level sparse attention

PIVOT (Proxy Indexing Via One full-prefix Traversal) is a training-free, drop-in replacement for the DeepSeek Sparse Attention (DSA) indexer that reduces computational redundancy by sharing one prefix scan across a group of nearby queries. Instead of scoring every preceding token for each query individually, PIVOT aggregates the group into a single proxy query to obtain a candidate set, from which top-k tokens are selected for each query.

arxiv arXiv cs.AI · 17d ago · 1 view

PoTRE introduces heterogeneous multi-agent framework for test-time reasoning

The authors introduce PoTRE (Poly-Topological Reasoning Ensembles), a framework designed to address the limitations of standard single-stream prompting in complex reasoning tasks. PoTRE decouples inference into four distinct agents: an Adversarial Refinement Agent, a Hierarchical strategic Planning Agent, a Spectrum Search Agent, and a Direct Chain Agent.