bibkey: chailloux2016parityoblivious authors: André Chailloux; Iordanis Kerenidis; Srijita Kundu; Jamie Sikora year: 2016 title: “Optimal bounds for parity-oblivious random access codes” doi: 10.1088/1367-2630/18/4/045003 url: https://arxiv.org/abs/1404.5153v3 claim: “The preparation-noncontextual n-bit parity-oblivious random-access-code bias is at most one over n.” strata_touched:

• D5/S3/QuantumContext/HanceTwentyFourCellThreshold license: citation-only triage: anchor

Verified locator

DOI: https://doi.org/10.1088/1367-2630/18/4/045003

Source: https://arxiv.org/abs/1404.5153v3

The preparation-contextuality discussion and the Spekkens et al. attribution, source CKKS_PO_arxiv_v3.tex, lines 369–372.

Prior upper obstruction

The sixteen half-sign preparations and the four coordinate questions form the four-bit parity-oblivious random-access-code obstruction. Its bias is lambda divided by two, and the noncontextual bias bound one over four gives lambda at most one half. This known obstruction does not by itself construct a model for all twenty-four preparations and all twelve sharp questions.