{"paper":{"title":"On-chip 1 TOPS Hyperdimensional Photonic Tensor Core using a WDM Silicon Photonic Coherent Crossbar","license":"http://creativecommons.org/licenses/by/4.0/","headline":"A silicon photonic crossbar achieves 0.96 TOPS for hyperdimensional tensor computations using time-space-wavelength multiplexing.","cross_cats":[],"primary_cat":"physics.optics","authors_text":"A. Tsakyridis, D. Lazovsky, I. Roumpos, K. Vyrsokinos, M. Moralis-Pegios, N. Pleros, S. Kovaios","submitted_at":"2026-05-13T09:15:03Z","abstract_excerpt":"We demonstrate an on-chip 0.96 TOPS hyperdimensional photonic tensor core by utilizing a time-spacewavelength multiplexed silicon photonic Crossbar (Xbar). The novel architecture relies on serializing the large matrix-vector or tensor-vector products by unfolding multiply and accumulation operations over time domain, while simultaneously distributing the computational workload over different spatial and wavelength channels. We experimentally demonstrate the operation of a 4-channel 2-input TSWDM Xbar that incorporates 56 GHz electroabsorption modulators (EAMs) and 4-channel integrated multiple"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"We demonstrate an on-chip 0.96 TOPS hyperdimensional photonic tensor core by utilizing a time-space-wavelength multiplexed silicon photonic Crossbar (Xbar).","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"The assumption that the demonstrated 4x2x1 unit performance and error rates will hold when scaling to larger arrays and higher channel counts without significant additional noise, crosstalk, or power penalties.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"An experimental 4-channel TSWDM silicon photonic crossbar achieves 0.96 TOPS for hyperdimensional tensor operations with 3.9% average error and 93.3% Iris accuracy at 10-30 GBd.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"A silicon photonic crossbar achieves 0.96 TOPS for hyperdimensional tensor computations using time-space-wavelength multiplexing.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"ab22afe3a49a256798bb66492ba2df46b4ccdc3c49974e777e6e0020118a0f33"},"source":{"id":"2605.13224","kind":"arxiv","version":1},"verdict":{"id":"f11dd665-dcee-4992-ab3d-0411336aafd8","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-14T18:52:41.629049Z","strongest_claim":"We demonstrate an on-chip 0.96 TOPS hyperdimensional photonic tensor core by utilizing a time-space-wavelength multiplexed silicon photonic Crossbar (Xbar).","one_line_summary":"An experimental 4-channel TSWDM silicon photonic crossbar achieves 0.96 TOPS for hyperdimensional tensor operations with 3.9% average error and 93.3% Iris accuracy at 10-30 GBd.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"The assumption that the demonstrated 4x2x1 unit performance and error rates will hold when scaling to larger arrays and higher channel counts without significant additional noise, crosstalk, or power penalties.","pith_extraction_headline":"A silicon photonic crossbar achieves 0.96 TOPS for hyperdimensional tensor computations using time-space-wavelength multiplexing."},"references":{"count":3,"sample":[{"doi":"10.1016/j.joule.2023.09.004","year":2023,"title":"The Growing Energy Footprint of Artificial Intelligence","work_id":"d9e9c97b-a398-4bb2-97b4-89e4514c57ae","ref_index":1,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"10.1038/s41566-020-00754-y","year":null,"title":"Pho- tonics for artificial intelligence and neuromorphic comput ing","work_id":"71bbdbe7-4edb-4c30-9c2c-0e1b80e36223","ref_index":2,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"10.48550/arxiv.2501.07917","year":2020,"title":"Roadmap on neuromorphic photonics","work_id":"d50ee3d8-f0ec-41de-95c0-0d3eb6befe10","ref_index":3,"cited_arxiv_id":"","is_internal_anchor":false}],"resolved_work":3,"snapshot_sha256":"977576dd79b445fd2b7ca76232b215a90d7b9ce764afabe653017e17c7ba2e68","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"}