Xinzhi Zhang 与 Zhao Song 完成 KLS 猜想证明Chinese only
微软研究员 Xinzhi Zhang 与合作者 Zhao Song 证明了提出于 1995 年的 KLS 猜想：对一类高维概率分布，平面切割与任意形状的最佳切割之间存在不随维度增长的固定界限。此前最佳结果仍允许界限随维度缓慢增长。研究者借助 AI 探索候选思路，并由人筛选和验证证明；另有独立证明及 Lean 4 形式化结果出现。
1 developments2 reports2 sources
Event developments 1
Xinzhi Zhang and Zhao Song prove the KLS conjecture
Microsoft researcher Xinzhi Zhang and collaborator Zhao Song have proved the KLS conjecture, proposed in 1995: for a broad class of high-dimensional probability distributions, the best flat cut remains within a fixed bound of the best cut of any shape, regardless of dimension. The previous best bound could still grow slowly with dimension. The researchers used AI to explore candidate approaches while directing and verifying the proof; separate proofs and a Lean 4 formalization have also appeared.
View source reports · 2 sources · 2 reports
- Microsoft Research · Xinzhi Zhang and Zhao Song prove the KLS conjecture
- Rohan Paul (@rohanpaul_ai) · Post says researchers used AI-assisted search to prove the KLS conjecture
All reports 2
Post says researchers used AI-assisted search to prove the KLS conjecture
A post says researchers directed a search in which AI agents explored roughly 100 research paths, while a researcher discarded weaker directions between rounds. The post credits the recursive method and error bound that completed the KLS proof to Song and Zhang.
1 more source reported this
- Microsoft Research
Xinzhi Zhang and Zhao Song prove the KLS conjecture
Microsoft researcher Xinzhi Zhang and collaborator Zhao Song have proved the KLS conjecture, proposed in 1995: for a broad class of high-dimensional probability distributions, the best flat cut remains within a fixed bound of the best cut of any shape, regardless of dimension. The previous best bound could still grow slowly with dimension. The researchers used AI to explore candidate approaches while directing and verifying the proof; separate proofs and a Lean 4 formalization have also appeared.
1 more source reported this
- Rohan Paul (@rohanpaul_ai)