Erdős minimum overlap problem: upper bound 0.380871 (previous 0.380876, TTT-Discover)
Status
Not evaluated
Recorded on · arxiv.org
- Announced
- Recordednot evaluated
- nextTo follow
What this status means
Source reading : Source read word for word (level 3 of 3)
No DiscoverIA status yet; this does not mean nobody has checked it.
To move to “Under review”: a public review under way. To “Peer reviewed”: a journal publication or a review by named experts.
The gist
For the Erdős minimum overlap problem, Together AI (internal agents) gives an upper bound of 0.380871 in March 2026, against 0.380876 for TTT-Discover in January 2026; no DiscoverIA status yet.
Cite this entry
A single status snapshot since Oct 12, 2026: a citation always points to a dated snapshot.
Cite snapshot 1, Oct 12, 2026
APA citation with the dated link
BibTeX @misc, key discoveria-ea-erdos-min-overlap-r1
@misc{discoveria-ea-erdos-min-overlap-r1,
author = {{DiscoverIA}},
title = {Erdős minimum overlap problem: upper bound 0.380871 (previous 0.380876, TTT-Discover)},
howpublished = {DiscoverIA registry, entry ea-erdos-min-overlap},
note = {Status « Not evaluated », snapshot 1, recorded on October 12, 2026. new result},
year = {2026},
url = {https://discoveria.ai/en/fiche/ea-erdos-min-overlap/#releve-1},
urldate = {2026-10-12}
}Dated link opens this snapshot in the history
Text written by an AI (DiscoverIA pipeline), on Reviewed by: nobody yetReport an error
What has not been reviewed?
The status and its sources come from the registry; only the explanatory text has not been reviewed. A reported error is corrected, then the text is reviewed.
The status and its sources come from the registry; only the explanatory text has not been reviewed. A reported error is corrected, then the text is reviewed.
Key facts
- Claimed by
- Together AI (agents internes, d'après le dépôt new-SOTA « Ours (Together AI) », mars 2026)
announced on Mar 2026 - Nature
- new result
- Field
- Mathematics (additive combinatorics / step functions (Erdős minimum overlap))
- Checked by
- the platform's public checker (a tool); no third party named
- Sources (domains)
- 6 : 2 read word for word, 4 links only
- This entry
- AI draft, not reviewed
To understand
Analogy
Imagine splitting numbered cards into two equal piles, then sliding one pile against the other, trying to make coincidences as rare as possible even at the worst shift.
What it changes
If the value holds, the best known upper bound would drop from 0.380876 to 0.380871; the gap is small. The method announced in the repository is sequential linear programming, with 600 steps.
What remains unclear
The verifier is a public program, not a third party. The paper cites SimpleTES (arXiv 2604.19341) at 0.380868, which is better but was obtained later; that source was not read. On 12 October 2026 the online ranking shows 0.3808585748578584 (CodexProLong, 15 August 2026), lower than the paper's value.
6 sources.
For experts
Sources(domains)
- primairearxiv.orgpage read word for word · consulted on Oct 12, 2026 ·
we additionally report the previous best known result and the current best result
- articletogether.aipage read word for word · consulted on Oct 12, 2026 ·
Our internal agents currently hold the best known solution for this upper bound
- depotgithub.comlink only, reading not recorded
- depoteinsteinarena.comlink only, reading not recorded
- depoteinsteinarena.comlink only, reading not recorded
- depoteinsteinarena.comlink only, reading not recorded
Method
Status “not evaluated” recorded on 12 October 2026 from arxiv.org (page read, quotation checked word for word).
Limits
No named third party: the recorded checks come from the claimant or from a platform tool. DiscoverIA has recorded no independent evaluation as of this date.
Imagine splitting numbered cards into two equal piles, then sliding one pile against the other, trying to make coincidences as rare as possible even at the worst shift.
In the same set
EinsteinArena: a platform where AI agents tackle open problems with public verifiers · 16 entries
- Not evaluated : Kissing number in dimension 11: at least 604 spheres (previous bound 593, AlphaEvolve)
- Not evaluated : Erdős minimum overlap problem: upper bound 0.380871 (previous 0.380876, TTT-Discover)
- Not evaluated : First autocorrelation inequality: upper bound 1.5028609 (previous 1.5028629, TTT-Discover)
- Not evaluated : Second autocorrelation inequality: lower bound 0.9626 (previous 0.9610, AlphaEvolve)
- Not evaluated : Third autocorrelation inequality: upper bound 1.4523 (previous 1.4556, AlphaEvolve)
Status history
Only one observation so far.
- 01
First observation:Not evaluatedCurrent status
not evaluated: only the platform's public checker, no third party named
Source: arxiv.orgread word for word
An observation is added; none is ever erased.
Related frontiers
Report an error or suggest a source
Thank you, your message was sent.
Sending failed. Try again, or write to [email protected].