{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:JJQHKR4JOTEW73PEGFZTYLSIFZ","short_pith_number":"pith:JJQHKR4J","schema_version":"1.0","canonical_sha256":"4a6075478974c96fede431733c2e482e56dc5652f02d1f5328a0f971d624dba2","source":{"kind":"arxiv","id":"2410.05037","version":1},"attestation_state":"computed","paper":{"title":"Improving Speaker Representations Using Contrastive Losses on Multi-scale Features","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["eess.AS"],"primary_cat":"cs.SD","authors_text":"Bhiksha Raj, Massa Baali, Rita Singh, Satvik Dixit","submitted_at":"2024-10-07T13:49:45Z","abstract_excerpt":"Speaker verification systems have seen significant advancements with the introduction of Multi-scale Feature Aggregation (MFA) architectures, such as MFA-Conformer and ECAPA-TDNN. These models leverage information from various network depths by concatenating intermediate feature maps before the pooling and projection layers, demonstrating that even shallower feature maps encode valuable speaker-specific information. Building upon this foundation, we propose a Multi-scale Feature Contrastive (MFCon) loss that directly enhances the quality of these intermediate representations. Our MFCon loss ap"},"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":"2410.05037","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.SD","submitted_at":"2024-10-07T13:49:45Z","cross_cats_sorted":["eess.AS"],"title_canon_sha256":"7d733011c0cbaafa00f3f27e2bfc8cc871dbb1b070c047f4d70955059fb2c0f5","abstract_canon_sha256":"afec510ff4c2a6ccb7cbc5ae58d55d0a2013faba00ef7e062a1a27b5367e9129"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:17:00.498988Z","signature_b64":"LzH4J6ptVGKYxjoTCiuVHuHX+m3pOjiTTLRtJfFwdpjNjUiZxjen5u0vCxDOcZxbJBWNAgh5WjADKO+A3BHKCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"4a6075478974c96fede431733c2e482e56dc5652f02d1f5328a0f971d624dba2","last_reissued_at":"2026-07-05T09:17:00.497639Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:17:00.497639Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Improving Speaker Representations Using Contrastive Losses on Multi-scale Features","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["eess.AS"],"primary_cat":"cs.SD","authors_text":"Bhiksha Raj, Massa Baali, Rita Singh, Satvik Dixit","submitted_at":"2024-10-07T13:49:45Z","abstract_excerpt":"Speaker verification systems have seen significant advancements with the introduction of Multi-scale Feature Aggregation (MFA) architectures, such as MFA-Conformer and ECAPA-TDNN. These models leverage information from various network depths by concatenating intermediate feature maps before the pooling and projection layers, demonstrating that even shallower feature maps encode valuable speaker-specific information. Building upon this foundation, we propose a Multi-scale Feature Contrastive (MFCon) loss that directly enhances the quality of these intermediate representations. Our MFCon loss ap"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.05037","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/2410.05037/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":"2410.05037","created_at":"2026-07-05T09:17:00.498479+00:00"},{"alias_kind":"arxiv_version","alias_value":"2410.05037v1","created_at":"2026-07-05T09:17:00.498479+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.05037","created_at":"2026-07-05T09:17:00.498479+00:00"},{"alias_kind":"pith_short_12","alias_value":"JJQHKR4JOTEW","created_at":"2026-07-05T09:17:00.498479+00:00"},{"alias_kind":"pith_short_16","alias_value":"JJQHKR4JOTEW73PE","created_at":"2026-07-05T09:17:00.498479+00:00"},{"alias_kind":"pith_short_8","alias_value":"JJQHKR4J","created_at":"2026-07-05T09:17:00.498479+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.21215","citing_title":"Speaker Identity in Non-Verbal Vocalizations: Conditional Distillation and Mixture of Experts Approach","ref_index":28,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/JJQHKR4JOTEW73PEGFZTYLSIFZ","json":"https://pith.science/pith/JJQHKR4JOTEW73PEGFZTYLSIFZ.json","graph_json":"https://pith.science/api/pith-number/JJQHKR4JOTEW73PEGFZTYLSIFZ/graph.json","events_json":"https://pith.science/api/pith-number/JJQHKR4JOTEW73PEGFZTYLSIFZ/events.json","paper":"https://pith.science/paper/JJQHKR4J"},"agent_actions":{"view_html":"https://pith.science/pith/JJQHKR4JOTEW73PEGFZTYLSIFZ","download_json":"https://pith.science/pith/JJQHKR4JOTEW73PEGFZTYLSIFZ.json","view_paper":"https://pith.science/paper/JJQHKR4J","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2410.05037&json=true","fetch_graph":"https://pith.science/api/pith-number/JJQHKR4JOTEW73PEGFZTYLSIFZ/graph.json","fetch_events":"https://pith.science/api/pith-number/JJQHKR4JOTEW73PEGFZTYLSIFZ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/JJQHKR4JOTEW73PEGFZTYLSIFZ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/JJQHKR4JOTEW73PEGFZTYLSIFZ/action/storage_attestation","attest_author":"https://pith.science/pith/JJQHKR4JOTEW73PEGFZTYLSIFZ/action/author_attestation","sign_citation":"https://pith.science/pith/JJQHKR4JOTEW73PEGFZTYLSIFZ/action/citation_signature","submit_replication":"https://pith.science/pith/JJQHKR4JOTEW73PEGFZTYLSIFZ/action/replication_record"}},"created_at":"2026-07-05T09:17:00.498479+00:00","updated_at":"2026-07-05T09:17:00.498479+00:00"}