{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:PE77VCYN2OQ4NY2DNVBEZVXDZN","short_pith_number":"pith:PE77VCYN","schema_version":"1.0","canonical_sha256":"793ffa8b0dd3a1c6e3436d424cd6e3cb55681528091a13a62a68dd99ddae286c","source":{"kind":"arxiv","id":"2310.13451","version":1},"attestation_state":"computed","paper":{"title":"Two-Stage Triplet Loss Training with Curriculum Augmentation for Audio-Visual Retrieval","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CV","cs.IR","cs.MM","eess.AS"],"primary_cat":"cs.SD","authors_text":"Donghuo Zeng, Kazushi Ikeda","submitted_at":"2023-10-20T12:35:54Z","abstract_excerpt":"The cross-modal retrieval model leverages the potential of triple loss optimization to learn robust embedding spaces. However, existing methods often train these models in a singular pass, overlooking the distinction between semi-hard and hard triples in the optimization process. The oversight of not distinguishing between semi-hard and hard triples leads to suboptimal model performance. In this paper, we introduce a novel approach rooted in curriculum learning to address this problem. We propose a two-stage training paradigm that guides the model's learning process from semi-hard to hard trip"},"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":"2310.13451","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.SD","submitted_at":"2023-10-20T12:35:54Z","cross_cats_sorted":["cs.CV","cs.IR","cs.MM","eess.AS"],"title_canon_sha256":"d260d05d3abbf440a79ae27c481a894e765c19b8e039a0130dde4c60ce0a8319","abstract_canon_sha256":"53db0a8a4b018f049aa71095591d4b339c475720ecf8521ac960b6ad26e0c77c"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:03:03.797569Z","signature_b64":"hCfoNrvvbeGo40il3RDWXy+YyuhtdvLZqqwlFiG+dwy6SHUYSdIQXIudPMYZf442eTn52XTrDOVfkPzwqfOlCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"793ffa8b0dd3a1c6e3436d424cd6e3cb55681528091a13a62a68dd99ddae286c","last_reissued_at":"2026-07-05T07:03:03.797111Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:03:03.797111Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Two-Stage Triplet Loss Training with Curriculum Augmentation for Audio-Visual Retrieval","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CV","cs.IR","cs.MM","eess.AS"],"primary_cat":"cs.SD","authors_text":"Donghuo Zeng, Kazushi Ikeda","submitted_at":"2023-10-20T12:35:54Z","abstract_excerpt":"The cross-modal retrieval model leverages the potential of triple loss optimization to learn robust embedding spaces. However, existing methods often train these models in a singular pass, overlooking the distinction between semi-hard and hard triples in the optimization process. The oversight of not distinguishing between semi-hard and hard triples leads to suboptimal model performance. In this paper, we introduce a novel approach rooted in curriculum learning to address this problem. We propose a two-stage training paradigm that guides the model's learning process from semi-hard to hard trip"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2310.13451","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/2310.13451/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":"2310.13451","created_at":"2026-07-05T07:03:03.797190+00:00"},{"alias_kind":"arxiv_version","alias_value":"2310.13451v1","created_at":"2026-07-05T07:03:03.797190+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2310.13451","created_at":"2026-07-05T07:03:03.797190+00:00"},{"alias_kind":"pith_short_12","alias_value":"PE77VCYN2OQ4","created_at":"2026-07-05T07:03:03.797190+00:00"},{"alias_kind":"pith_short_16","alias_value":"PE77VCYN2OQ4NY2D","created_at":"2026-07-05T07:03:03.797190+00:00"},{"alias_kind":"pith_short_8","alias_value":"PE77VCYN","created_at":"2026-07-05T07:03:03.797190+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2501.09608","citing_title":"Metric Learning with Progressive Self-Distillation for Audio-Visual Embedding Learning","ref_index":25,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/PE77VCYN2OQ4NY2DNVBEZVXDZN","json":"https://pith.science/pith/PE77VCYN2OQ4NY2DNVBEZVXDZN.json","graph_json":"https://pith.science/api/pith-number/PE77VCYN2OQ4NY2DNVBEZVXDZN/graph.json","events_json":"https://pith.science/api/pith-number/PE77VCYN2OQ4NY2DNVBEZVXDZN/events.json","paper":"https://pith.science/paper/PE77VCYN"},"agent_actions":{"view_html":"https://pith.science/pith/PE77VCYN2OQ4NY2DNVBEZVXDZN","download_json":"https://pith.science/pith/PE77VCYN2OQ4NY2DNVBEZVXDZN.json","view_paper":"https://pith.science/paper/PE77VCYN","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2310.13451&json=true","fetch_graph":"https://pith.science/api/pith-number/PE77VCYN2OQ4NY2DNVBEZVXDZN/graph.json","fetch_events":"https://pith.science/api/pith-number/PE77VCYN2OQ4NY2DNVBEZVXDZN/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/PE77VCYN2OQ4NY2DNVBEZVXDZN/action/timestamp_anchor","attest_storage":"https://pith.science/pith/PE77VCYN2OQ4NY2DNVBEZVXDZN/action/storage_attestation","attest_author":"https://pith.science/pith/PE77VCYN2OQ4NY2DNVBEZVXDZN/action/author_attestation","sign_citation":"https://pith.science/pith/PE77VCYN2OQ4NY2DNVBEZVXDZN/action/citation_signature","submit_replication":"https://pith.science/pith/PE77VCYN2OQ4NY2DNVBEZVXDZN/action/replication_record"}},"created_at":"2026-07-05T07:03:03.797190+00:00","updated_at":"2026-07-05T07:03:03.797190+00:00"}