Unverified50% confidenceFactExact time
A preprint describes a deterministic algorithm that can exactly count the number of satisfying assignments of a certain class of KCNF formulas in subexponential time, which if valid would challenge the Exponential Time Hypothesis (ETH).
1
Sources
50%
Confidence
Long-term
Relevance
10/30/2026
First Seen
Sources
OpenAI's Claimed Refutation of ETH Sparks Debate, Meta Bets Big on AI Infrastructure
youtubeIamNobot10/27/2026
Related Entities
Related Claims
UnverifiedThe model's results on the Kakeya conjecture, ETH and QCD are currently still at the stage of requiring external verification.70% similarUnverifiedThe study's methodology pre-registered hypotheses, thresholds, and analysis methods, used deterministic regular expressions for scoring rather than a large model as judge, and calculated confidence intervals via bill-level clustered bootstrapping.67% similarUnverifiedThe paper proves an exploration lower bound that holds for any predictable step sizes, and combines it with a statistical lower bound to yield matching mean-squared rates up to logarithmic factors.66% similarUnverifiedIn simulation experiments over 270 grid points and under non-Gaussian seed distributions, the closed-form formula was validated within the relevant design range (from five runs per method).64% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1116018API
curl https://kongchang.com/api/v1/knowledge/claims/1116018MCP
get_claim(id=1116018)