{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2020:TPUAZTQH4NDELELCP2IF6CWXML","short_pith_number":"pith:TPUAZTQH","schema_version":"1.0","canonical_sha256":"9be80cce07e3464591627e905f0ad762c4ab90a092fc7f7ae94b1732e7cca0a2","source":{"kind":"arxiv","id":"2002.06275","version":1},"attestation_state":"computed","paper":{"title":"TwinBERT: Distilling Knowledge to Twin-Structured BERT Models for Efficient Retrieval","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.IR","authors_text":"Jian Jiao, Ruofei Zhang, Wenhao Lu","submitted_at":"2020-02-14T22:44:36Z","abstract_excerpt":"Pre-trained language models like BERT have achieved great success in a wide variety of NLP tasks, while the superior performance comes with high demand in computational resources, which hinders the application in low-latency IR systems. We present TwinBERT model for effective and efficient retrieval, which has twin-structured BERT-like encoders to represent query and document respectively and a crossing layer to combine the embeddings and produce a similarity score. Different from BERT, where the two input sentences are concatenated and encoded together, TwinBERT decouples them during encoding"},"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":"2002.06275","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.IR","submitted_at":"2020-02-14T22:44:36Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"949210331308fb3a0d42e766a7f169f4a4230ed0f2eb5a309920be92c39c48a0","abstract_canon_sha256":"d84b6fe68fdb5f82f60dbd2892d26269bb513b9d94a8cedfb052203e4b37fc0e"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T00:41:04.860930Z","signature_b64":"bBrP4wPsa7AF6ysLXIWg/28ff4ZidXZBN/0hpSO0Q6uHASpwFQ6zH700EFfIW1pH7xEzuifyn+yNoKc4tEWeDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9be80cce07e3464591627e905f0ad762c4ab90a092fc7f7ae94b1732e7cca0a2","last_reissued_at":"2026-07-05T00:41:04.860579Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T00:41:04.860579Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"TwinBERT: Distilling Knowledge to Twin-Structured BERT Models for Efficient Retrieval","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.IR","authors_text":"Jian Jiao, Ruofei Zhang, Wenhao Lu","submitted_at":"2020-02-14T22:44:36Z","abstract_excerpt":"Pre-trained language models like BERT have achieved great success in a wide variety of NLP tasks, while the superior performance comes with high demand in computational resources, which hinders the application in low-latency IR systems. We present TwinBERT model for effective and efficient retrieval, which has twin-structured BERT-like encoders to represent query and document respectively and a crossing layer to combine the embeddings and produce a similarity score. Different from BERT, where the two input sentences are concatenated and encoded together, TwinBERT decouples them during encoding"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2002.06275","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/2002.06275/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":"2002.06275","created_at":"2026-07-05T00:41:04.860643+00:00"},{"alias_kind":"arxiv_version","alias_value":"2002.06275v1","created_at":"2026-07-05T00:41:04.860643+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2002.06275","created_at":"2026-07-05T00:41:04.860643+00:00"},{"alias_kind":"pith_short_12","alias_value":"TPUAZTQH4NDE","created_at":"2026-07-05T00:41:04.860643+00:00"},{"alias_kind":"pith_short_16","alias_value":"TPUAZTQH4NDELELC","created_at":"2026-07-05T00:41:04.860643+00:00"},{"alias_kind":"pith_short_8","alias_value":"TPUAZTQH","created_at":"2026-07-05T00:41:04.860643+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2502.09089","citing_title":"Semantic Ads Retrieval at Walmart eCommerce with Language Models Progressively Trained on Multiple Knowledge Domains","ref_index":12,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/TPUAZTQH4NDELELCP2IF6CWXML","json":"https://pith.science/pith/TPUAZTQH4NDELELCP2IF6CWXML.json","graph_json":"https://pith.science/api/pith-number/TPUAZTQH4NDELELCP2IF6CWXML/graph.json","events_json":"https://pith.science/api/pith-number/TPUAZTQH4NDELELCP2IF6CWXML/events.json","paper":"https://pith.science/paper/TPUAZTQH"},"agent_actions":{"view_html":"https://pith.science/pith/TPUAZTQH4NDELELCP2IF6CWXML","download_json":"https://pith.science/pith/TPUAZTQH4NDELELCP2IF6CWXML.json","view_paper":"https://pith.science/paper/TPUAZTQH","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2002.06275&json=true","fetch_graph":"https://pith.science/api/pith-number/TPUAZTQH4NDELELCP2IF6CWXML/graph.json","fetch_events":"https://pith.science/api/pith-number/TPUAZTQH4NDELELCP2IF6CWXML/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/TPUAZTQH4NDELELCP2IF6CWXML/action/timestamp_anchor","attest_storage":"https://pith.science/pith/TPUAZTQH4NDELELCP2IF6CWXML/action/storage_attestation","attest_author":"https://pith.science/pith/TPUAZTQH4NDELELCP2IF6CWXML/action/author_attestation","sign_citation":"https://pith.science/pith/TPUAZTQH4NDELELCP2IF6CWXML/action/citation_signature","submit_replication":"https://pith.science/pith/TPUAZTQH4NDELELCP2IF6CWXML/action/replication_record"}},"created_at":"2026-07-05T00:41:04.860643+00:00","updated_at":"2026-07-05T00:41:04.860643+00:00"}