{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:VN4XDXLDZF5QSVL5YRVXRBWJ4F","short_pith_number":"pith:VN4XDXLD","schema_version":"1.0","canonical_sha256":"ab7971dd63c97b09557dc46b7886c9e1487cb12e02facd7a41123b9c9f40a2a7","source":{"kind":"arxiv","id":"2407.07479","version":1},"attestation_state":"computed","paper":{"title":"How to Make Cross Encoder a Good Teacher for Efficient Image-Text Retrieval?","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Bing Li, Chunfeng Yuan, Junfu Pu, Weiming Hu, Xiaojuan Qi, Ying Shan, Yuxin Chen, Zhongang Qi, Ziqi Zhang, Zongyang Ma","submitted_at":"2024-07-10T09:10:01Z","abstract_excerpt":"Dominant dual-encoder models enable efficient image-text retrieval but suffer from limited accuracy while the cross-encoder models offer higher accuracy at the expense of efficiency. Distilling cross-modality matching knowledge from cross-encoder to dual-encoder provides a natural approach to harness their strengths. Thus we investigate the following valuable question: how to make cross-encoder a good teacher for dual-encoder? Our findings are threefold:(1) Cross-modal similarity score distribution of cross-encoder is more concentrated while the result of dual-encoder is nearly normal making v"},"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":"2407.07479","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2024-07-10T09:10:01Z","cross_cats_sorted":[],"title_canon_sha256":"17f29d3def91b868525f14f6701dfe62707918ccf07af19c4d7c3a094243cb0b","abstract_canon_sha256":"9d8b3197f8d32ec79cb2739709b9fc09533f82f89bfc864e0a31a3714e43affc"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:42:12.554539Z","signature_b64":"374Ex0CeNbeil51sNyeA8SuTjhdLCKDDfpWiav3pcUp84UaFu42l6CEvcULAjPsVUP53/TigyL25BX87JxlqDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ab7971dd63c97b09557dc46b7886c9e1487cb12e02facd7a41123b9c9f40a2a7","last_reissued_at":"2026-07-05T08:42:12.554068Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:42:12.554068Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"How to Make Cross Encoder a Good Teacher for Efficient Image-Text Retrieval?","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Bing Li, Chunfeng Yuan, Junfu Pu, Weiming Hu, Xiaojuan Qi, Ying Shan, Yuxin Chen, Zhongang Qi, Ziqi Zhang, Zongyang Ma","submitted_at":"2024-07-10T09:10:01Z","abstract_excerpt":"Dominant dual-encoder models enable efficient image-text retrieval but suffer from limited accuracy while the cross-encoder models offer higher accuracy at the expense of efficiency. Distilling cross-modality matching knowledge from cross-encoder to dual-encoder provides a natural approach to harness their strengths. Thus we investigate the following valuable question: how to make cross-encoder a good teacher for dual-encoder? Our findings are threefold:(1) Cross-modal similarity score distribution of cross-encoder is more concentrated while the result of dual-encoder is nearly normal making v"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2407.07479","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/2407.07479/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":"2407.07479","created_at":"2026-07-05T08:42:12.554126+00:00"},{"alias_kind":"arxiv_version","alias_value":"2407.07479v1","created_at":"2026-07-05T08:42:12.554126+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2407.07479","created_at":"2026-07-05T08:42:12.554126+00:00"},{"alias_kind":"pith_short_12","alias_value":"VN4XDXLDZF5Q","created_at":"2026-07-05T08:42:12.554126+00:00"},{"alias_kind":"pith_short_16","alias_value":"VN4XDXLDZF5QSVL5","created_at":"2026-07-05T08:42:12.554126+00:00"},{"alias_kind":"pith_short_8","alias_value":"VN4XDXLD","created_at":"2026-07-05T08:42:12.554126+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/VN4XDXLDZF5QSVL5YRVXRBWJ4F","json":"https://pith.science/pith/VN4XDXLDZF5QSVL5YRVXRBWJ4F.json","graph_json":"https://pith.science/api/pith-number/VN4XDXLDZF5QSVL5YRVXRBWJ4F/graph.json","events_json":"https://pith.science/api/pith-number/VN4XDXLDZF5QSVL5YRVXRBWJ4F/events.json","paper":"https://pith.science/paper/VN4XDXLD"},"agent_actions":{"view_html":"https://pith.science/pith/VN4XDXLDZF5QSVL5YRVXRBWJ4F","download_json":"https://pith.science/pith/VN4XDXLDZF5QSVL5YRVXRBWJ4F.json","view_paper":"https://pith.science/paper/VN4XDXLD","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2407.07479&json=true","fetch_graph":"https://pith.science/api/pith-number/VN4XDXLDZF5QSVL5YRVXRBWJ4F/graph.json","fetch_events":"https://pith.science/api/pith-number/VN4XDXLDZF5QSVL5YRVXRBWJ4F/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/VN4XDXLDZF5QSVL5YRVXRBWJ4F/action/timestamp_anchor","attest_storage":"https://pith.science/pith/VN4XDXLDZF5QSVL5YRVXRBWJ4F/action/storage_attestation","attest_author":"https://pith.science/pith/VN4XDXLDZF5QSVL5YRVXRBWJ4F/action/author_attestation","sign_citation":"https://pith.science/pith/VN4XDXLDZF5QSVL5YRVXRBWJ4F/action/citation_signature","submit_replication":"https://pith.science/pith/VN4XDXLDZF5QSVL5YRVXRBWJ4F/action/replication_record"}},"created_at":"2026-07-05T08:42:12.554126+00:00","updated_at":"2026-07-05T08:42:12.554126+00:00"}