This preprint (arXiv:2604.14323, submitted April 2026) by Hela Mhiri and collaborators extends boson sampling theory into the saturated regime where the number of optical modes scales linearly or sub-quadratically with photon number. Unlike most prior work that assumed a "dilute" limit (modes >> n² for n photons) where collisions are rare, the authors tackle the crowded conditions closest to current photonic experiments.

Using representation-theoretic techniques from the symmetric group and unitary group actions on Fock space, the team derives closed-form expressions for the second moments of any particle-number-preserving bosonic observable. These are expressed via Hilbert-Schmidt norms of projections onto irreducible representation subspaces; the symmetry structure yields compact analytical formulas that avoid exponential sums. The methodology is purely theoretical—exact algebraic derivations with no numerical simulation or empirical data, and therefore no sample size.

A core result is an anti-concentration bound on Fock-state output probabilities that holds beyond the dilute regime. Anti-concentration (showing that no single outcome dominates the distribution) is a crucial ingredient in complexity proofs that relate approximate sampling to the hardness of computing permanents.

Previous coverage and even many theoretical papers, including follow-ups to the foundational 2011 Aaronson-Arkhipov work ("The computational complexity of linear optics," STOC 2011, arXiv:1011.3245), largely sidestepped this saturated regime or relied on conjectures. Experimental milestone papers such as Zhong et al. (Science, 2020, arXiv:2012.01625) demonstrating Gaussian boson sampling with up to 76 photons operated squarely in the regime where collisions matter, yet lacked full theoretical backing for hardness. This preprint closes that gap.

The analysis reveals patterns seen across quantum advantage proposals: initial experimental claims (random circuit sampling, boson sampling) outpace theory, then mathematical tools—here Schur-Weyl duality and moment calculations—catch up to legitimize them. What the original abstract understates is the practical implication: experimentalists no longer need to artificially dilute their setups to retain provable hardness, lowering a key barrier to scalable photonic quantum advantage.

Limitations are explicit. The work assumes ideal, lossless linear optics and perfect single-photon sources and detectors; real devices suffer from photon loss, mode mismatch, and partial distinguishability. As a preprint it has not completed peer review. Still, by synthesizing representation theory with complexity results published in the last two years, it supplies a missing theoretical pillar. The advance strengthens confidence that photonic systems can deliver verifiable quantum advantage without retreating to theoretically convenient but experimentally artificial parameter regimes.