VibeMathedMath problems solved by AI

Improved Approximation Ratios for Multiway Cut

New upper and lower bounds on the approximation ratio achievable for Multiway Cut via large mixtures of new and old rounding schemes for the CKR relaxation, advancing the ratio ladder that has run since Călinescu-Karloff-Rabani (1998).

Result
Proved(see note)
Status
Partial result
AI contribution
AI-assisted
Method
Computation
Field
Approximation algorithms
Posed by
Year posed
Years open
Solved
2026-03-30
Model
ChatGPT
Vendor
OpenAI
Collaborators
Joshua Brakensiek, Neng Huang, Aaron Potechin, Uri Zwick
Verification
Unreviewed
Publication
Preprint
Significance
15 / 100
Disclosed cost
Wikipedia
No dedicated article

What was actually shown

Record bounds on the ratio; the exact approximability of Multiway Cut remains open.

What the AI did

"We acknowledge help from ChatGPT while writing the code for discovering new rounding schemes and while preparing some of the plots. We emphasize that we did not use ChatGPT or any other LLM model while writing our verification code."

Source

arXiv

Discussion