AI Solves Two-Year-Old ICML Math Conjecture in Minutes
abeirami · x · 2026-08-02
AI researcher @abeirami shared a breakthrough test of LLM reasoning capabilities. Two years ago, their team proposed Conjecture 4.4 in their ICML'25 paper regarding a tighter upper bound on the KL divergence of best-of-n sampling relative to the reference model, which they struggled to solve for weeks.
However, during a recent lunch break, they fed the conjecture to Fable 5 and GPT-5.6 Sol. Surprisingly, both models produced clear, self-contained mathematical proofs within minutes. The author noted their only contribution was a few prompts to extract the key non-trivial step as a separate lemma, with the core reasoning entirely generated by the AI.
Related event: AI Proves Two-Year-Old Math Conjecture in Minutes(3 posts)→
More from Models
- Open-weight models now beat Claude Sonnet 5 as gap closes fast — LegacyRemaster · 2026-10-16
- GYRO quant of Qwen3.8-Flash-Next tested: near Q3 quality at much smaller size — PlasticRevenue4601 · 2026-10-16
- Open source is so back: dozens of open models dropped within 48 hours — beffjezos · 2026-10-16
- Gemini users hit usage caps on 3.1 Pro and 4 Argon, resets not until Oct 16 — R3d-Gr33n-Blu3 · 2026-10-16
- Gemini desktop app can't accept images — an image model that can't take images — Rebl · 2026-10-16
- Apollo Research: GLM-5.3 unprompted reasons about its actions becoming training data — _lewtun · 2026-10-16