Single-exponential temperature cost for one-dimensional Gibbs preparation
Yuxuan Zhang — version 1, 22 October 2026
Affiliations: EPFL first; Princeton University second, as shown in the manuscript.

main.tex is the complete standalone paper; manuscript.pdf is its reviewed export.
Build: pdflatex main.tex, twice. Standard LaTeX packages are used.

Reproduction: install requirements.txt in a Python environment, then run python reproduce.py.
The driver verifies release-manifest.json and runs the selected unmodified checkers in a fresh temporary copy.
reproduction-results records their output. Finite controls do not prove universal analytic statements.
The seven earlier papers retain a small shared evidence snapshot so original relative checker paths remain valid.
Third-party full-text snapshots are not republished; the paper and comparison records give primary-source links.
Historical snapshot/full-candidate manifests describe the larger original local freeze and may reference files outside this selected release.
release-manifest.json is the inventory for this package. Historical proof-stage status labels remain unchanged;
the complete paper and dated independent review record are the current assessment.

Exact candidate: M05-v1. The power of the logarithm grows as O(x); a temperature-uniform exponent and the suggested quadratic temperature exponent are not proved. The result is state preparation, with discarded randomness included in the output state. Arbitrarily long binary inputs retain the displayed polynomial bit-cost factor.

Internal independent-agent proof and source review found no specific unresolved blocker. This is not external peer review,
formal proof-assistant certification, or an exhaustive priority guarantee. Website release only; no archive or submission is implied.
