{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:AMNCMNMEJBMQALEGQDOOL7VG47","short_pith_number":"pith:AMNCMNME","schema_version":"1.0","canonical_sha256":"031a2635844859002c8680dce5fea6e7e96234b9aff063310194a3cdc38ad1b9","source":{"kind":"arxiv","id":"2408.16672","version":4},"attestation_state":"computed","paper":{"title":"Jina-ColBERT-v2: A General-Purpose Multilingual Late Interaction Retriever","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.IR","authors_text":"Andreas Koukounas, Bo Wang, Georgios Mastrapas, Han Xiao, Isabelle Mohr, Michael G\\\"unther, Mohammad Kalim Akram, Nan Wang, Rohan Jha, Saba Sturua","submitted_at":"2024-08-29T16:21:00Z","abstract_excerpt":"Multi-vector dense models, such as ColBERT, have proven highly effective in information retrieval. ColBERT's late interaction scoring approximates the joint query-document attention seen in cross-encoders while maintaining inference efficiency closer to traditional dense retrieval models, thanks to its bi-encoder architecture and recent optimizations in indexing and search. In this work we propose a number of incremental improvements to the ColBERT model architecture and training pipeline, using methods shown to work in the more mature single-vector embedding model training paradigm, particula"},"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":"2408.16672","kind":"arxiv","version":4},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.IR","submitted_at":"2024-08-29T16:21:00Z","cross_cats_sorted":["cs.AI","cs.CL"],"title_canon_sha256":"6bec2ef547dff4a8a4fb7ac575a0fdf705bd80f4141ddf3c0a4d176da209ede1","abstract_canon_sha256":"f4def17e77296114da1ee32521b30e2d190993127f596b5d9675593d6cd2d233"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:07:02.603194Z","signature_b64":"244Rxy7LF68e8Pz2IbH05LCuJ7hBs0aBiW2nWoLjDK+AZFbJ5OGl9GHqkMwv1E/FAAtGJksnb6qfVAUVKpRLBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"031a2635844859002c8680dce5fea6e7e96234b9aff063310194a3cdc38ad1b9","last_reissued_at":"2026-07-05T09:07:02.602730Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:07:02.602730Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Jina-ColBERT-v2: A General-Purpose Multilingual Late Interaction Retriever","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.IR","authors_text":"Andreas Koukounas, Bo Wang, Georgios Mastrapas, Han Xiao, Isabelle Mohr, Michael G\\\"unther, Mohammad Kalim Akram, Nan Wang, Rohan Jha, Saba Sturua","submitted_at":"2024-08-29T16:21:00Z","abstract_excerpt":"Multi-vector dense models, such as ColBERT, have proven highly effective in information retrieval. ColBERT's late interaction scoring approximates the joint query-document attention seen in cross-encoders while maintaining inference efficiency closer to traditional dense retrieval models, thanks to its bi-encoder architecture and recent optimizations in indexing and search. In this work we propose a number of incremental improvements to the ColBERT model architecture and training pipeline, using methods shown to work in the more mature single-vector embedding model training paradigm, particula"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2408.16672","kind":"arxiv","version":4},"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/2408.16672/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":"2408.16672","created_at":"2026-07-05T09:07:02.602789+00:00"},{"alias_kind":"arxiv_version","alias_value":"2408.16672v4","created_at":"2026-07-05T09:07:02.602789+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2408.16672","created_at":"2026-07-05T09:07:02.602789+00:00"},{"alias_kind":"pith_short_12","alias_value":"AMNCMNMEJBMQ","created_at":"2026-07-05T09:07:02.602789+00:00"},{"alias_kind":"pith_short_16","alias_value":"AMNCMNMEJBMQALEG","created_at":"2026-07-05T09:07:02.602789+00:00"},{"alias_kind":"pith_short_8","alias_value":"AMNCMNME","created_at":"2026-07-05T09:07:02.602789+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2508.01408","citing_title":"Artificial Intelligence and Misinformation in Art: Can Vision Language Models Judge the Hand or the Machine Behind the Canvas?","ref_index":67,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/AMNCMNMEJBMQALEGQDOOL7VG47","json":"https://pith.science/pith/AMNCMNMEJBMQALEGQDOOL7VG47.json","graph_json":"https://pith.science/api/pith-number/AMNCMNMEJBMQALEGQDOOL7VG47/graph.json","events_json":"https://pith.science/api/pith-number/AMNCMNMEJBMQALEGQDOOL7VG47/events.json","paper":"https://pith.science/paper/AMNCMNME"},"agent_actions":{"view_html":"https://pith.science/pith/AMNCMNMEJBMQALEGQDOOL7VG47","download_json":"https://pith.science/pith/AMNCMNMEJBMQALEGQDOOL7VG47.json","view_paper":"https://pith.science/paper/AMNCMNME","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2408.16672&json=true","fetch_graph":"https://pith.science/api/pith-number/AMNCMNMEJBMQALEGQDOOL7VG47/graph.json","fetch_events":"https://pith.science/api/pith-number/AMNCMNMEJBMQALEGQDOOL7VG47/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/AMNCMNMEJBMQALEGQDOOL7VG47/action/timestamp_anchor","attest_storage":"https://pith.science/pith/AMNCMNMEJBMQALEGQDOOL7VG47/action/storage_attestation","attest_author":"https://pith.science/pith/AMNCMNMEJBMQALEGQDOOL7VG47/action/author_attestation","sign_citation":"https://pith.science/pith/AMNCMNMEJBMQALEGQDOOL7VG47/action/citation_signature","submit_replication":"https://pith.science/pith/AMNCMNMEJBMQALEGQDOOL7VG47/action/replication_record"}},"created_at":"2026-07-05T09:07:02.602789+00:00","updated_at":"2026-07-05T09:07:02.602789+00:00"}