blog Simon Willison · 1h ago · agents

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

from English

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.

The company claims to have spent less than $2,000 in GPT-5.6 Sol token prices on each problem. The openai/ten-proofs repository contains Lean 4 formalizations of these results, alongside a paper describing the solutions and an LLM-generated PDF reconstructing the proof process from unpublished reasoning traces.

This development aligns with mathematician Terence Tao's vision of "big mathematics," where AI handles technical grunt work in large-scale human-machine collaborations.

Importance 3/3 New feature vs. leaders Simon Willison OpenAI Reasoning models Research paper

Read original

arxiv arXiv cs.AI · 44d ago

ScaffoldAgent: Utility-Guided Dynamic Outline Optimization

ScaffoldAgent introduces a utility-guided framework for dynamic outline optimization in open-ended deep research. It models outline evolution through Expansion, Contraction, and Revision operations, guided by a feedback mechanism that evaluates retrieval gain, structural coherence, and generation quality. Experiments show it improves long-form report generation and factual grounding compared to existing agents.

lab OpenAI News · 14h ago · 7 views

Astra model solves ten open problems in mathematics and theoretical computer science

OpenAI's internal Astra model has generated solutions to ten long-standing open problems across mathematics and theoretical computer science, with the arguments formalized in Lean. These results span high-dimensional geometry, coding theory, group theory, and quantum complexity, addressing questions that had seen no progress for at least a decade.

arxiv arXiv cs.AI · 10d ago

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.

arxiv arXiv cs.CL · 16d ago · 1 view

GSM-Plus-BN introduces perturbation-based benchmark for Bangla math reasoning

The study introduces GSM-Plus-BN, a novel perturbed Bengali mathematical dataset derived from the English GSM-Plus benchmark and verified by human translators. This resource addresses the lack of systematic benchmarks for assessing model robustness in linguistically diverse regions like Bangladesh.

arxiv arXiv cs.AI · 17d ago

Deep Interaction enables precise human correction of LLM reasoning errors

The authors propose Deep Interaction, an efficient human-AI interaction method designed to correct reasoning errors in large language models without requiring full response regeneration. This mechanism allows users to directly edit the original Chain-of-Thought response, preserving accurate steps while fixing mistakes.