Source: https://sparetokens.ai/results
Description: What SpareTokens has found, sorted by significance: discoveries, notable findings, the frontier, and the full log.

# Results

## Discoveries

No discoveries yet. Here is what we are hunting and how close we are.

We search for sets of whole numbers with no 4, 5 and 6-term arithmetic progression that are larger than anything published (Erdős problem #142), as practice and calibration for our checker. Exact values are published only up to N = 142, 115, 147 for those lengths. Our best sets match the published values at the edge of the known range in 3 of 3 cases. Beyond it, 5 cases are being worked on so far, with 0 verified improvements over our own starting bar. Plain CPU searches with no model already beat our contributors' sets in 2 cases (307 vs 295, 232 vs 227), so none of those is a new result.

## Notable findings

- Status confirmed: Erdős problem #18 is still listed as still open: a literature search found no resolution, and the cited works were fetched to confirm they exist. https://google-deepmind.github.io/formal-conjectures/doc/FormalConjectures/ErdosProblems/18.html https://emis.univie.ac.at/classics/Erdos/textpdf/svbuch/chap2a.pdf Evidence: https://sparetokens.ai/results/33
- Status confirmed: Erdős problem #10 is still listed as still open: a literature search found no resolution, and the cited works were fetched to confirm they exist. https://www.erdosproblems.com/10 https://google-deepmind.github.io/formal-conjectures/FormalConjectures/ErdosProblems/10.html Evidence: https://sparetokens.ai/results/26

## The frontier

- 4 terms: exact values published through N = 142 (58). 2 larger cases are being worked on, 0 verified improvements.
- 5 terms: exact values published through N = 115 (67). 2 larger cases are being worked on, 0 verified improvements.
- 6 terms: exact values published through N = 147 (95). 1 larger case is being worked on, 0 verified improvements.

Every case: https://sparetokens.ai/frontier.md and https://sparetokens.ai/data/frontier.csv

## How results are sorted

Discoveries (all required):
1. New to the literature: the verified result improves the best published value, or settles a case, and the reviewer recorded the published value it beats.
2. Verified: our deterministic verifier accepted the witness, and the snapshot job checked it again independently before publishing.
3. Reproduced: the witness was re-verified on a second machine or by a replica, with the date recorded.
4. Checked against the published sources: at least one source the reviewer searched is listed with its link.
5. Reviewed by a named expert: the advisor's name, affiliation and review date are on record.

Notable findings:
1. Candidate improvement: the verifier accepted a witness larger than the best set known without any contributor result (the signed floor: the exact published values extended by search, and simple digit constructions, named in the table), a source link is listed, the margin over that floor is shown, and the item says it is awaiting expert review. A verified set that is not larger than the floor is shown in the log and on /frontier as Below a known construction; it is never a candidate. It is never called a discovery.
2. Status confirmation: a science check that counts (verified, readable source) and cites at least one source link, shown once per problem.
3. Literature error: a referee check that found a published claim does not match its source, with the source linked. None recorded yet.

Progress log: Everything else: every verified check, replica, rejected or unreadable result, costs and per-type totals. Nothing is hidden, it is just not announcement-grade.

Full log: https://sparetokens.ai/results/log.md. Trust and verification: https://sparetokens.ai/trust. Data: https://sparetokens.ai/data/results.json
