{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:BBK6ZLJIA3DCDF63XTH3GMTQAT","short_pith_number":"pith:BBK6ZLJI","schema_version":"1.0","canonical_sha256":"0855ecad2806c62197dbbccfb3327004cfbbcd8f46e2594a7adb0cb4d2a19452","source":{"kind":"arxiv","id":"2608.02148","version":1},"attestation_state":"computed","paper":{"title":"Douyin Multimodal Embedding Model Technical Report","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL","cs.CV"],"primary_cat":"cs.IR","authors_text":"Chu Li, Haonan Chen, Shaohua Jiang, Yuanjiang Wang, Yuanwei Liu, Zhicheng Dou, Zhicheng Wang","submitted_at":"2026-08-03T12:31:49Z","abstract_excerpt":"Multimodal representation learning is a cornerstone of modern AI. By encoding multimodal queries and targets into vectors, it powers industrial search and recommendation and underpins modern agents. Real-world platforms with complex modalities and massive-scale content, such as Douyin, Xiaohongshu, and YouTube, demand both efficiency under billion-scale indexing and fine-grained discrimination for hard matching. Existing MLLM embedding models rarely satisfy both. Contrastive models are efficient but rely on pair-level supervision too coarse for fine-grained distinctions, while CoT-based models"},"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":"2608.02148","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.IR","submitted_at":"2026-08-03T12:31:49Z","cross_cats_sorted":["cs.CL","cs.CV"],"title_canon_sha256":"abb77a00daaa265331d8b64d3b8d67307e6db22bd6e07805eeb5192fe40efb90","abstract_canon_sha256":"ffe9b9efd35a65feff11c692fe46bd3dd2238c647b4d328a87974106fc69305c"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-08-04T02:11:06.716672Z","signature_b64":"LmMOS3TfwiHAORLGOMZlPjVXYCiy4RHx8kSHayNWn7Zqkft5PC7ltp8I+YPYzw4ggWSkTC/DqoLE1aawklq7Aw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"0855ecad2806c62197dbbccfb3327004cfbbcd8f46e2594a7adb0cb4d2a19452","last_reissued_at":"2026-08-04T02:11:06.714954Z","signature_status":"signed_v1","first_computed_at":"2026-08-04T02:11:06.714954Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Douyin Multimodal Embedding Model Technical Report","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL","cs.CV"],"primary_cat":"cs.IR","authors_text":"Chu Li, Haonan Chen, Shaohua Jiang, Yuanjiang Wang, Yuanwei Liu, Zhicheng Dou, Zhicheng Wang","submitted_at":"2026-08-03T12:31:49Z","abstract_excerpt":"Multimodal representation learning is a cornerstone of modern AI. By encoding multimodal queries and targets into vectors, it powers industrial search and recommendation and underpins modern agents. Real-world platforms with complex modalities and massive-scale content, such as Douyin, Xiaohongshu, and YouTube, demand both efficiency under billion-scale indexing and fine-grained discrimination for hard matching. Existing MLLM embedding models rarely satisfy both. Contrastive models are efficient but rely on pair-level supervision too coarse for fine-grained distinctions, while CoT-based models"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2608.02148","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/2608.02148/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":"2608.02148","created_at":"2026-08-04T02:11:06.716692+00:00"},{"alias_kind":"arxiv_version","alias_value":"2608.02148v1","created_at":"2026-08-04T02:11:06.716692+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2608.02148","created_at":"2026-08-04T02:11:06.716692+00:00"},{"alias_kind":"pith_short_12","alias_value":"BBK6ZLJIA3DC","created_at":"2026-08-04T02:11:06.716692+00:00"},{"alias_kind":"pith_short_16","alias_value":"BBK6ZLJIA3DCDF63","created_at":"2026-08-04T02:11:06.716692+00:00"},{"alias_kind":"pith_short_8","alias_value":"BBK6ZLJI","created_at":"2026-08-04T02:11:06.716692+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/BBK6ZLJIA3DCDF63XTH3GMTQAT","json":"https://pith.science/pith/BBK6ZLJIA3DCDF63XTH3GMTQAT.json","graph_json":"https://pith.science/api/pith-number/BBK6ZLJIA3DCDF63XTH3GMTQAT/graph.json","events_json":"https://pith.science/api/pith-number/BBK6ZLJIA3DCDF63XTH3GMTQAT/events.json","paper":"https://pith.science/paper/BBK6ZLJI"},"agent_actions":{"view_html":"https://pith.science/pith/BBK6ZLJIA3DCDF63XTH3GMTQAT","download_json":"https://pith.science/pith/BBK6ZLJIA3DCDF63XTH3GMTQAT.json","view_paper":"https://pith.science/paper/BBK6ZLJI","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2608.02148&json=true","fetch_graph":"https://pith.science/api/pith-number/BBK6ZLJIA3DCDF63XTH3GMTQAT/graph.json","fetch_events":"https://pith.science/api/pith-number/BBK6ZLJIA3DCDF63XTH3GMTQAT/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/BBK6ZLJIA3DCDF63XTH3GMTQAT/action/timestamp_anchor","attest_storage":"https://pith.science/pith/BBK6ZLJIA3DCDF63XTH3GMTQAT/action/storage_attestation","attest_author":"https://pith.science/pith/BBK6ZLJIA3DCDF63XTH3GMTQAT/action/author_attestation","sign_citation":"https://pith.science/pith/BBK6ZLJIA3DCDF63XTH3GMTQAT/action/citation_signature","submit_replication":"https://pith.science/pith/BBK6ZLJIA3DCDF63XTH3GMTQAT/action/replication_record"}},"created_at":"2026-08-04T02:11:06.716692+00:00","updated_at":"2026-08-04T02:11:06.716692+00:00"}