What was already known
Previous work settled parts of the Courtade–Kumar conjecture under restrictions on the function or the noise. Other full proofs appeared in the same period.
AI involvement
The authors say generative AI produced the single-bit proof and its detailed derivations. They supplied high-level direction, feedback and final review.
- Generative AI disclosureGenerative AI disclosure
Importance
This is a separate entropy-flow proof of the full one-bit problem, including the previously difficult unbalanced case.
Technical verification
The manuscript and Lean project identify the exported one-bit theorem and its independent-noise setup.
Formal checking
Checked the one-bit bound for all Boolean functions under independent binary symmetric noise.
- Checker
- Lean 4 kernel
- Scope
- Mahdavifar and Beirami give an independent proof that a coordinate is the most informative one-bit summary of a uniform Boolean string under independent binary symmetric noise. The proof covers biased Boolean functions and the full noise range.
- Basis
- Reported by the linked checker
- Lean development for the one-bit theoremLean development for the one-bit theorem
Human understanding
The authors describe direction, feedback and review. The disclosure says the detailed one-bit derivations came from generative AI and leaves a human reconstruction undocumented.
Other useful sources
Correction and publication history
Reviewed the primary result and its documented AI and checking scope.
Read this published revision