{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:GXEJZ3GFE3BHVHHBDDQPKJVGL7","short_pith_number":"pith:GXEJZ3GF","schema_version":"1.0","canonical_sha256":"35c89cecc526c27a9ce118e0f526a65fcfd192969f32be3d1b53f2541b002770","source":{"kind":"arxiv","id":"2210.05156","version":2},"attestation_state":"computed","paper":{"title":"Task-Aware Specialization for Efficient and Robust Dense Retrieval for Open-Domain Question Answering","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Hao Cheng, Hao Fang, Jianfeng Gao, XiaoDong Liu","submitted_at":"2022-10-11T05:33:25Z","abstract_excerpt":"Given its effectiveness on knowledge-intensive natural language processing tasks, dense retrieval models have become increasingly popular. Specifically, the de-facto architecture for open-domain question answering uses two isomorphic encoders that are initialized from the same pretrained model but separately parameterized for questions and passages. This bi-encoder architecture is parameter-inefficient in that there is no parameter sharing between encoders. Further, recent studies show that such dense retrievers underperform BM25 in various settings. We thus propose a new architecture, Task-aw"},"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":"2210.05156","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2022-10-11T05:33:25Z","cross_cats_sorted":[],"title_canon_sha256":"d0a8723f4348e479e92917a771fa1890ee111cf49d50ab1b167640e871f444bf","abstract_canon_sha256":"09a1e1a65d6bd2052295198cff976e2a0df1ba44d010dcd820533121eb7b18cc"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:12:37.821865Z","signature_b64":"focqnbHa9AKEwzQ2BeAbgdaY6RkUx16o9Mtu+RH4nKuDeH42455FQYTcfbdIacP6SNuiLTix2TEEighjQMUDCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"35c89cecc526c27a9ce118e0f526a65fcfd192969f32be3d1b53f2541b002770","last_reissued_at":"2026-07-05T06:12:37.821400Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:12:37.821400Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Task-Aware Specialization for Efficient and Robust Dense Retrieval for Open-Domain Question Answering","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Hao Cheng, Hao Fang, Jianfeng Gao, XiaoDong Liu","submitted_at":"2022-10-11T05:33:25Z","abstract_excerpt":"Given its effectiveness on knowledge-intensive natural language processing tasks, dense retrieval models have become increasingly popular. Specifically, the de-facto architecture for open-domain question answering uses two isomorphic encoders that are initialized from the same pretrained model but separately parameterized for questions and passages. This bi-encoder architecture is parameter-inefficient in that there is no parameter sharing between encoders. Further, recent studies show that such dense retrievers underperform BM25 in various settings. We thus propose a new architecture, Task-aw"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2210.05156","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/2210.05156/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":"2210.05156","created_at":"2026-07-05T06:12:37.821455+00:00"},{"alias_kind":"arxiv_version","alias_value":"2210.05156v2","created_at":"2026-07-05T06:12:37.821455+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2210.05156","created_at":"2026-07-05T06:12:37.821455+00:00"},{"alias_kind":"pith_short_12","alias_value":"GXEJZ3GFE3BH","created_at":"2026-07-05T06:12:37.821455+00:00"},{"alias_kind":"pith_short_16","alias_value":"GXEJZ3GFE3BHVHHB","created_at":"2026-07-05T06:12:37.821455+00:00"},{"alias_kind":"pith_short_8","alias_value":"GXEJZ3GF","created_at":"2026-07-05T06:12:37.821455+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2508.11779","citing_title":"A Multi-Task Evaluation of LLMs' Processing of Academic Text Input","ref_index":11,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/GXEJZ3GFE3BHVHHBDDQPKJVGL7","json":"https://pith.science/pith/GXEJZ3GFE3BHVHHBDDQPKJVGL7.json","graph_json":"https://pith.science/api/pith-number/GXEJZ3GFE3BHVHHBDDQPKJVGL7/graph.json","events_json":"https://pith.science/api/pith-number/GXEJZ3GFE3BHVHHBDDQPKJVGL7/events.json","paper":"https://pith.science/paper/GXEJZ3GF"},"agent_actions":{"view_html":"https://pith.science/pith/GXEJZ3GFE3BHVHHBDDQPKJVGL7","download_json":"https://pith.science/pith/GXEJZ3GFE3BHVHHBDDQPKJVGL7.json","view_paper":"https://pith.science/paper/GXEJZ3GF","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2210.05156&json=true","fetch_graph":"https://pith.science/api/pith-number/GXEJZ3GFE3BHVHHBDDQPKJVGL7/graph.json","fetch_events":"https://pith.science/api/pith-number/GXEJZ3GFE3BHVHHBDDQPKJVGL7/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/GXEJZ3GFE3BHVHHBDDQPKJVGL7/action/timestamp_anchor","attest_storage":"https://pith.science/pith/GXEJZ3GFE3BHVHHBDDQPKJVGL7/action/storage_attestation","attest_author":"https://pith.science/pith/GXEJZ3GFE3BHVHHBDDQPKJVGL7/action/author_attestation","sign_citation":"https://pith.science/pith/GXEJZ3GFE3BHVHHBDDQPKJVGL7/action/citation_signature","submit_replication":"https://pith.science/pith/GXEJZ3GFE3BHVHHBDDQPKJVGL7/action/replication_record"}},"created_at":"2026-07-05T06:12:37.821455+00:00","updated_at":"2026-07-05T06:12:37.821455+00:00"}