,
Shachar Lovett
,
Kunal Mittal
Creative Commons Attribution 4.0 International license
We prove that for any 3-player game G, whose query distribution has the same support as the GHZ game (i.e., all x,y,z ∈ {0,1} satisfying x+y+z = 0 (mod 2)), the value of the n-fold parallel repetition of G decays exponentially fast:
val(G^{⊗ n}) ≤ exp(-n^c)
for all sufficiently large n, where c > 0 is an absolute constant.
We also prove a concentration bound for the parallel repetition of the GHZ game: For any constant ε > 0, the probability that the players win at least a (3/4+ε) fraction of the n coordinates is at most exp(-n^c), where c = c(ε) > 0 is a constant.
In both settings, our work exponentially improves upon the previous best known bounds which were only polynomially small, i.e., of the order n^{-Ω(1)}. Our key technical tool is the notion of algebraic spreadness adapted from the breakthrough work of Kelley and Meka (FOCS '23) on sets free of 3-term progressions.
@InProceedings{liu_et_al:LIPIcs.CCC.2026.6,
author = {Liu, Yang P. and Lovett, Shachar and Mittal, Kunal},
title = {{Improved Parallel Repetition for GHZ-Supported Games via Spreadness}},
booktitle = {41st Computational Complexity Conference (CCC 2026)},
pages = {6:1--6:21},
series = {Leibniz International Proceedings in Informatics (LIPIcs)},
ISBN = {978-3-95977-437-6},
ISSN = {1868-8969},
year = {2026},
volume = {383},
editor = {Moshkovitz, Dana},
publisher = {Schloss Dagstuhl -- Leibniz-Zentrum f{\"u}r Informatik},
address = {Dagstuhl, Germany},
URL = {https://drops.dagstuhl.de/entities/document/10.4230/LIPIcs.CCC.2026.6},
URN = {urn:nbn:de:0030-drops-270486},
doi = {10.4230/LIPIcs.CCC.2026.6},
annote = {Keywords: Parallel Repetition, GHZ Game, Algebraic Spreadness}
}