{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:SR2SLFWWIY7EBDUH2FKCN33TYK","short_pith_number":"pith:SR2SLFWW","schema_version":"1.0","canonical_sha256":"94752596d6463e408e87d15426ef73c296ba57a8fe10ab49f4b98912e4bd47e4","source":{"kind":"arxiv","id":"2112.08723","version":2},"attestation_state":"computed","paper":{"title":"Distilled Dual-Encoder Model for Vision-Language Understanding","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CV"],"primary_cat":"cs.CL","authors_text":"Bing Qin, Furu Wei, Haichao Zhu, Ming Liu, Wenhui Wang, Zekun Wang","submitted_at":"2021-12-16T09:21:18Z","abstract_excerpt":"We propose a cross-modal attention distillation framework to train a dual-encoder model for vision-language understanding tasks, such as visual reasoning and visual question answering. Dual-encoder models have a faster inference speed than fusion-encoder models and enable the pre-computation of images and text during inference. However, the shallow interaction module used in dual-encoder models is insufficient to handle complex vision-language understanding tasks. In order to learn deep interactions of images and text, we introduce cross-modal attention distillation, which uses the image-to-te"},"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":"2112.08723","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2021-12-16T09:21:18Z","cross_cats_sorted":["cs.CV"],"title_canon_sha256":"b712911025a98cf8b30f2a9bdebb8cdf07e9e3fc6a25559845eea582721ff378","abstract_canon_sha256":"41ca8a2d00681a3f497946c62fec5493e954e0420603f0162e8e837a7deff631"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:06:53.758499Z","signature_b64":"LWEgL262X85re3733qTvjJ/fNjFjfaagXpm2OsCguW9OR5tLGgKUbE1naY+JB8rPrDh30AOobRd4i5+Z3g1/Cg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"94752596d6463e408e87d15426ef73c296ba57a8fe10ab49f4b98912e4bd47e4","last_reissued_at":"2026-07-05T05:06:53.758005Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:06:53.758005Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Distilled Dual-Encoder Model for Vision-Language Understanding","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CV"],"primary_cat":"cs.CL","authors_text":"Bing Qin, Furu Wei, Haichao Zhu, Ming Liu, Wenhui Wang, Zekun Wang","submitted_at":"2021-12-16T09:21:18Z","abstract_excerpt":"We propose a cross-modal attention distillation framework to train a dual-encoder model for vision-language understanding tasks, such as visual reasoning and visual question answering. Dual-encoder models have a faster inference speed than fusion-encoder models and enable the pre-computation of images and text during inference. However, the shallow interaction module used in dual-encoder models is insufficient to handle complex vision-language understanding tasks. In order to learn deep interactions of images and text, we introduce cross-modal attention distillation, which uses the image-to-te"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2112.08723","kind":"arxiv","version":2},"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/2112.08723/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":"2112.08723","created_at":"2026-07-05T05:06:53.758081+00:00"},{"alias_kind":"arxiv_version","alias_value":"2112.08723v2","created_at":"2026-07-05T05:06:53.758081+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2112.08723","created_at":"2026-07-05T05:06:53.758081+00:00"},{"alias_kind":"pith_short_12","alias_value":"SR2SLFWWIY7E","created_at":"2026-07-05T05:06:53.758081+00:00"},{"alias_kind":"pith_short_16","alias_value":"SR2SLFWWIY7EBDUH","created_at":"2026-07-05T05:06:53.758081+00:00"},{"alias_kind":"pith_short_8","alias_value":"SR2SLFWW","created_at":"2026-07-05T05:06:53.758081+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2506.00479","citing_title":"EffiVLM-BENCH: A Comprehensive Benchmark for Evaluating Training-Free Acceleration in Large Vision-Language Models","ref_index":60,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/SR2SLFWWIY7EBDUH2FKCN33TYK","json":"https://pith.science/pith/SR2SLFWWIY7EBDUH2FKCN33TYK.json","graph_json":"https://pith.science/api/pith-number/SR2SLFWWIY7EBDUH2FKCN33TYK/graph.json","events_json":"https://pith.science/api/pith-number/SR2SLFWWIY7EBDUH2FKCN33TYK/events.json","paper":"https://pith.science/paper/SR2SLFWW"},"agent_actions":{"view_html":"https://pith.science/pith/SR2SLFWWIY7EBDUH2FKCN33TYK","download_json":"https://pith.science/pith/SR2SLFWWIY7EBDUH2FKCN33TYK.json","view_paper":"https://pith.science/paper/SR2SLFWW","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2112.08723&json=true","fetch_graph":"https://pith.science/api/pith-number/SR2SLFWWIY7EBDUH2FKCN33TYK/graph.json","fetch_events":"https://pith.science/api/pith-number/SR2SLFWWIY7EBDUH2FKCN33TYK/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/SR2SLFWWIY7EBDUH2FKCN33TYK/action/timestamp_anchor","attest_storage":"https://pith.science/pith/SR2SLFWWIY7EBDUH2FKCN33TYK/action/storage_attestation","attest_author":"https://pith.science/pith/SR2SLFWWIY7EBDUH2FKCN33TYK/action/author_attestation","sign_citation":"https://pith.science/pith/SR2SLFWWIY7EBDUH2FKCN33TYK/action/citation_signature","submit_replication":"https://pith.science/pith/SR2SLFWWIY7EBDUH2FKCN33TYK/action/replication_record"}},"created_at":"2026-07-05T05:06:53.758081+00:00","updated_at":"2026-07-05T05:06:53.758081+00:00"}