FAR Framework Shifts AI Math Research from Solving Problems to Finding Them
dan_fried · x · 2026-08-19
Traditional AI for math focuses on solving given problems, but mathematical discovery also involves deciding which problems are worth solving. A new paper introduces FAR (Find, Attempt, Recommend), a paradigm to scale this process.
How it works:
- Takes a broad research direction as input.
- Find: Scans literature for open conjectures.
- Attempt: Attempts to resolve them at scale.
- Recommend: Filters promising conjecture-resolution pairs for human expert review.
Impact: A pilot run on combinatorics literature surfaced potential resolutions to hundreds of open problems, demonstrating AI's potential to identify, not just solve, mathematical challenges.
Related event: FAR Framework Enables AI Mathematical Discovery, Solving 1977 Conjecture(3 posts)→
More from Research
- MIT PhD Sharut Gupta joins DeepMind; Isola hails her representational alignment defense talk — phillip_isola · 2026-10-16
- 36 primitives, just 10 parameters each: a remarkably precise T-pose body fit — letandrewcook · 2026-10-16
- Interp researcher pushes back: if you can't simulate a neuron, you haven't explained it — nabla_theta · 2026-10-16
- MLP-Splatting (ECCV 2026) code released: MLPs as light-field primitives instead of Gaussians — AjdDavison · 2026-10-16
- Four-page solution to subcubic APSP builds intuition for Anthropic-AI-aided breakthrough — gautamcgoel · 2026-10-16
- ATAS: Zhejiang-led team builds world-state memory framework for consistent agentic video generation — CSProfKGD · 2026-10-16