FAR Framework: AI Filters Math Conjectures Worth Human Review
rohanpaul_ai · x · 2026-08-31
As AI gets better at math, the harder problem is deciding which problems deserve compute and expert attention. The FAR (Finding and Ranking) framework shifts the focus from "solve this" to "find what's worth solving."
- Workflow: Starts with a research direction, searches literature for open problems, attempts them at scale, and filters outputs so experts only see the most promising cases.
- Results: In a combinatorics pilot, it narrowed 51,110 papers to 4,717 open conjectures, produced 1,050 new resolutions, passed 598 through automated judging, and recommended 77 for expert review. A manual check of 15 artifacts found all mathematically correct.
Related event: FAR Framework Helps AI Rank Which Math Conjectures Deserve Attention(2 posts)→
More from Research
- Interpretability debate: one interpretable polysemantic neuron doesn't prove all are — aran_nayebi · 2026-10-16
- Optuna creator Akiba: when 2 hyperparameters cost $20M, what's left for BO/HPO — Stefania_druga · 2026-10-16
- AI will be superhuman at science regardless, and Europe can't opt out — teortaxesTex · 2026-10-16
- SERO augments long-horizon agentic RL with executable oracle continuations — Shanghai-AI-Laboratory · 2026-10-16
- Tsinghua and Amap propose OmniVerifier-C1, a graded verifier beating binary judges — Bowei Liu · 2026-10-16
- Alibaba open-sources MAI-Gym, an all-in-one GUI agent eval and RL training platform — Tongyi-MAI · 2026-10-16