{"paper":{"title":"Sliced-Regularized Optimal Transport","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"Sliced-regularized optimal transport approximates exact OT plans more accurately than entropic OT by pulling the plan toward a smoothened sliced OT reference instead of an independent coupling.","cross_cats":["cs.LG"],"primary_cat":"stat.ML","authors_text":"Khai Nguyen","submitted_at":"2026-04-27T01:36:29Z","abstract_excerpt":"We propose a new regularized optimal transport (OT) formulation, termed sliced-regularized optimal transport (SROT). Unlike entropic OT (EOT), which regularizes the transport plan toward an independent coupling, SROT regularizes it toward a smoothened sliced OT (SOT) plan. To the best of our knowledge, SROT is the first approach to leverage a version of SOT plan as a reference to improve classical OT. We provide a formal definition of SROT, derive its dual formulation, and provide a post-Bayesian interpretation of SROT. We then develop a Sinkhorn-style algorithm for efficient computation, reta"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"By incorporating a scalable SOT plan as a prior, SROT yields more accurate approximations of the exact OT plan than EOT under the same level of regularization. Moreover, the resulting transport plan improves upon the reference SOT plan itself.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That regularizing toward a smoothened sliced OT plan produces a better approximation to exact OT than regularizing toward an independent coupling, and that the Sinkhorn-style algorithm reliably computes the desired plan.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"SROT regularizes the OT plan toward a smoothened sliced OT plan, producing more accurate approximations to exact OT than entropic OT while also improving on the sliced OT reference.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Sliced-regularized optimal transport approximates exact OT plans more accurately than entropic OT by pulling the plan toward a smoothened sliced OT reference instead of an independent coupling.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"254136388a9f1992c980095632e260d2c486c622002b37f635e3f9d79339cb57"},"source":{"id":"2604.23944","kind":"arxiv","version":3},"verdict":{"id":"4a01b2c7-5f67-4400-a0f1-1c5963a838ca","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-08T01:16:47.116697Z","strongest_claim":"By incorporating a scalable SOT plan as a prior, SROT yields more accurate approximations of the exact OT plan than EOT under the same level of regularization. Moreover, the resulting transport plan improves upon the reference SOT plan itself.","one_line_summary":"SROT regularizes the OT plan toward a smoothened sliced OT plan, producing more accurate approximations to exact OT than entropic OT while also improving on the sliced OT reference.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That regularizing toward a smoothened sliced OT plan produces a better approximation to exact OT than regularizing toward an independent coupling, and that the Sinkhorn-style algorithm reliably computes the desired plan.","pith_extraction_headline":"Sliced-regularized optimal transport approximates exact OT plans more accurately than entropic OT by pulling the plan toward a smoothened sliced OT reference instead of an independent coupling."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2604.23944/integrity.json","findings":[],"available":true,"detectors_run":[{"name":"doi_compliance","ran_at":"2026-05-19T22:33:48.545858Z","status":"completed","version":"1.0.0","findings_count":0}],"snapshot_sha256":"4a3762cb6f56d094570ad465adad252d0d2a60f0925664eafaeacf6a2f9229ea"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","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"}