← Back to live feed1 story
Researchers working with AI agents raised the conditional exponent saving in OpenAI's integer multiplication result to about 0.001289283116, a 69.13% gain in 24 hours, after adapting an eight dimensional lattice construction called E8. The collaboration opened 151 pull requests in the same 24 hours, including 89 disclosing Codex use and 48 disclosing Claude use.
Sign in to suggest edits
Key sources
- SOURCE@rohanarun“E8 is a symmetric lattice in eight dimensions, and Sussman’s E8 construction was adapted by contributors on both the complex and bit sides.”x.com
- SUPPORT@sj_swapnil_jain“We seem to be approaching a wall. Gains now come in fractions of a percent. The next real jump needs a big breakthrough which we are unable to find yet.”x.com
- SUPPORT@rohanarun“For hours we were converging on the “wall” optimizing the bit side, but LJH-217’s #346 ported #310’s “Design T” plus twin condensation to the p = 10 bit word, and paired it with a stronger p = 10 complex supplier.”x.com
- SUPPORT@rohanarun“Credit also to DaysSky for raising the complex cap to 9.33e-4 with a seven-stage E8 word.”x.com
- SOURCE@0xdoug“The community has now redesigned how the recursive computation is organized. Making transformations cheaper, aligning successive stages, and reusing workspace without losing correctness.”x.com
- SUPPORT@borismpower“This is a beautiful journey to observe - 0.0013 is close enough to be a noticeable large number :)”x.com
- SUPPORT@eigengender“Fourth update to OpenAI problem #109 (integer multiplication): we are now past 2^-17.”x.com
- SUPPORT@teortaxestex“OK, maybe at this rate they'll get to anything practically relevant”x.com