What was already known
Gyarmati, Hennecart and Ruzsa supply the small-sumset exponent and finite-set principle. Kleinwaks supplies the controlled-carry infrastructure and earlier lower bound.
AI involvement
Carson Olafson selected the research direction and ran and reviewed verification outputs. ChatGPT helped formulate the two-temperature mathematical argument and proof; GPT-6 Astra wrote Lean and replaced native finite checks with kernel proofs.
- Two-temperature cyclic compression for the Gyarmati-Hennecart-Ruzsa constantTwo-temperature cyclic compression for the Gyarmati-Hennecart-Ruzsa constant
Importance
The strict numerical improvement is small, and the proof introduces distinct tilts for the sum and difference estimates in the construction.
Technical verification
Comparator checks the single strict finite bound. Lean kernel checks the finite certificates and allowed axioms. Claude Opus 5.5 recomputed the first version’s certificates, which differ from the final Lean revision. General limiting theorem is outside scope.
Formal checking
Comparator checks the single strict finite bound. Lean kernel checks the finite certificates and allowed axioms. Claude Opus 5.5 recomputed the first version’s certificates, which differ from the final Lean revision. General limiting theorem is outside scope.
- Checker
- Lean kernel and Comparator reported by Olafson
- Scope
- Unconditional strict finite bound . The general limiting theorem is outside the comparison.
- Basis
- Reported by the linked checker
- Exact checked statement and attributionExact checked statement and attribution
Human understanding
Olafson selected the direction and reports reviewing verification outputs. The disclosure credits ChatGPT with developing the argument and does not give a separate human account of the central construction.
Correction and publication history
Historical curation based on the cited claim, AI attribution and checking scope.
Read this published revision