{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:O65XJBVQC2L6WHVQNYGJBCQLLK","short_pith_number":"pith:O65XJBVQ","schema_version":"1.0","canonical_sha256":"77bb7486b01697eb1eb06e0c908a0b5a82b1450a8c929a9f38efe86108c23c13","source":{"kind":"arxiv","id":"2310.13025","version":1},"attestation_state":"computed","paper":{"title":"Powerset multi-class cross entropy loss for neural speaker diarization","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CL","cs.NE","eess.AS"],"primary_cat":"cs.SD","authors_text":"Alexis Plaquet (IRIT-SAMoVA), Herv\\'e Bredin (IRIT-SAMoVA)","submitted_at":"2023-10-19T06:51:43Z","abstract_excerpt":"Since its introduction in 2019, the whole end-to-end neural diarization (EEND) line of work has been addressing speaker diarization as a frame-wise multi-label classification problem with permutation-invariant training. Despite EEND showing great promise, a few recent works took a step back and studied the possible combination of (local) supervised EEND diarization with (global) unsupervised clustering. Yet, these hybrid contributions did not question the original multi-label formulation. We propose to switch from multi-label (where any two speakers can be active at the same time) to powerset "},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2310.13025","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.SD","submitted_at":"2023-10-19T06:51:43Z","cross_cats_sorted":["cs.AI","cs.CL","cs.NE","eess.AS"],"title_canon_sha256":"015a9eccc573b0cc600bbd5b2065f7bc8420799fabc8c75c76834918d378037c","abstract_canon_sha256":"3f2299d6ac4f8c6294043957b94730f52ff1ac44e3814540f084852e33486577"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:02:55.991766Z","signature_b64":"5y9O3hpeqqq2+OHZGGDKrO3ICqWMVozoAfFi91YnzrSdTj0ZVSQSVpadg/bojkCfXOzwGOeyeIHx1NKWHdVVAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"77bb7486b01697eb1eb06e0c908a0b5a82b1450a8c929a9f38efe86108c23c13","last_reissued_at":"2026-07-05T07:02:55.991292Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:02:55.991292Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Powerset multi-class cross entropy loss for neural speaker diarization","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CL","cs.NE","eess.AS"],"primary_cat":"cs.SD","authors_text":"Alexis Plaquet (IRIT-SAMoVA), Herv\\'e Bredin (IRIT-SAMoVA)","submitted_at":"2023-10-19T06:51:43Z","abstract_excerpt":"Since its introduction in 2019, the whole end-to-end neural diarization (EEND) line of work has been addressing speaker diarization as a frame-wise multi-label classification problem with permutation-invariant training. Despite EEND showing great promise, a few recent works took a step back and studied the possible combination of (local) supervised EEND diarization with (global) unsupervised clustering. Yet, these hybrid contributions did not question the original multi-label formulation. We propose to switch from multi-label (where any two speakers can be active at the same time) to powerset "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2310.13025","kind":"arxiv","version":1},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2310.13025/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"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"},"aliases":[{"alias_kind":"arxiv","alias_value":"2310.13025","created_at":"2026-07-05T07:02:55.991360+00:00"},{"alias_kind":"arxiv_version","alias_value":"2310.13025v1","created_at":"2026-07-05T07:02:55.991360+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2310.13025","created_at":"2026-07-05T07:02:55.991360+00:00"},{"alias_kind":"pith_short_12","alias_value":"O65XJBVQC2L6","created_at":"2026-07-05T07:02:55.991360+00:00"},{"alias_kind":"pith_short_16","alias_value":"O65XJBVQC2L6WHVQ","created_at":"2026-07-05T07:02:55.991360+00:00"},{"alias_kind":"pith_short_8","alias_value":"O65XJBVQ","created_at":"2026-07-05T07:02:55.991360+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2506.02610","citing_title":"Speaker Diarization with Overlapping Community Detection Using Graph Attention Networks and Label Propagation Algorithm","ref_index":32,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/O65XJBVQC2L6WHVQNYGJBCQLLK","json":"https://pith.science/pith/O65XJBVQC2L6WHVQNYGJBCQLLK.json","graph_json":"https://pith.science/api/pith-number/O65XJBVQC2L6WHVQNYGJBCQLLK/graph.json","events_json":"https://pith.science/api/pith-number/O65XJBVQC2L6WHVQNYGJBCQLLK/events.json","paper":"https://pith.science/paper/O65XJBVQ"},"agent_actions":{"view_html":"https://pith.science/pith/O65XJBVQC2L6WHVQNYGJBCQLLK","download_json":"https://pith.science/pith/O65XJBVQC2L6WHVQNYGJBCQLLK.json","view_paper":"https://pith.science/paper/O65XJBVQ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2310.13025&json=true","fetch_graph":"https://pith.science/api/pith-number/O65XJBVQC2L6WHVQNYGJBCQLLK/graph.json","fetch_events":"https://pith.science/api/pith-number/O65XJBVQC2L6WHVQNYGJBCQLLK/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/O65XJBVQC2L6WHVQNYGJBCQLLK/action/timestamp_anchor","attest_storage":"https://pith.science/pith/O65XJBVQC2L6WHVQNYGJBCQLLK/action/storage_attestation","attest_author":"https://pith.science/pith/O65XJBVQC2L6WHVQNYGJBCQLLK/action/author_attestation","sign_citation":"https://pith.science/pith/O65XJBVQC2L6WHVQNYGJBCQLLK/action/citation_signature","submit_replication":"https://pith.science/pith/O65XJBVQC2L6WHVQNYGJBCQLLK/action/replication_record"}},"created_at":"2026-07-05T07:02:55.991360+00:00","updated_at":"2026-07-05T07:02:55.991360+00:00"}