{"id":"arxiv:2610.34411","kind":"preprint","title":"Bosonic Loss Codes with Bounded Occupation","authors":["Avraham Kreindel","Aryeh Lev Zabokritskiy"],"author_count":2,"year":2026,"published":"2026-10-25","venue":null,"cited_by":0,"doi":null,"arxiv_id":"2610.34411","nct_id":null,"source_url":"https://arxiv.org/abs/2610.34411","archive_url":"https://space-kitty.com/papers/arxiv:2610.34411","updated":"2026-10-25","pmid":null,"pmcid":null,"openalex_id":"W7223959249","pdf_url":"https://arxiv.org/pdf/2610.34411","abstract":"We study Fock-state codes on $q$ bosonic modes, with total excitation $N$ and at most $B$ excitations per mode, that exactly correct every pattern of at most $δN$ photon losses, for $q$ proportional to $N$. For every fixed cap $B$ and every ratio $q/N>1/B$, we prove explicit achievable rates, positive for all sufficiently small $δ>0$, in logical qubits per excitation. At $B=2$ and $q=N$, the rate is positive for every $δ<0.1355$ and is about $0.281$ at $δ=0.1$, compared with the previously established exact-code qutrit rate of about $0.275$. We also give an explicit upper bound on the dimension of every such exact code, including degenerate codes. For the pure-loss channel truncated to at most $δN$ losses and renormalized, we obtain approximate codes with exponentially small worst-case error. When the loss probability is positive and at most $1/(B+1)$, they improve our positive exact-code rate bounds; at $B=2$, $q=N$, and loss probability $0.1$, they remain positive up to $δ\\approx0.158$.","abstract_license":"CC0-1.0 (arXiv descriptive metadata)","license":null,"is_open_access":false,"retracted":false,"peer_reviewed":false,"categories":["quant-ph"],"details":null,"provenance":"arxiv","attribution":"Abstract: arXiv descriptive metadata (CC0). Thank you to arXiv for use of its open access interoperability. This service was not reviewed or approved by, nor does it necessarily express or reflect the policies or opinions of, arXiv.","topics":[{"slug":"quantum-error-correction","name":"Quantum Error Correction & Fault Tolerance","domain_slug":"quantum-computing"}],"cited_in_dossiers":[]}