{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:HBXEWCWNOMZWWIZIKPRLIZ7BS5","short_pith_number":"pith:HBXEWCWN","schema_version":"1.0","canonical_sha256":"386e4b0acd73336b232853e2b467e197500928817d12731e511b5940f4820594","source":{"kind":"arxiv","id":"2101.00117","version":1},"attestation_state":"computed","paper":{"title":"Multi-task Retrieval for Knowledge-Intensive Tasks","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Barlas O\\u{g}uz, Fabio Petroni, Gargi Ghosh, Jean Maillard, Veselin Stoyanov, Vladimir Karpukhin, Wen-tau Yih","submitted_at":"2021-01-01T00:16:34Z","abstract_excerpt":"Retrieving relevant contexts from a large corpus is a crucial step for tasks such as open-domain question answering and fact checking. Although neural retrieval outperforms traditional methods like tf-idf and BM25, its performance degrades considerably when applied to out-of-domain data.\n  Driven by the question of whether a neural retrieval model can be universal and perform robustly on a wide variety of problems, we propose a multi-task trained model. Our approach not only outperforms previous methods in the few-shot setting, but also rivals specialised neural retrievers, even when in-domain"},"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":"2101.00117","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.CL","submitted_at":"2021-01-01T00:16:34Z","cross_cats_sorted":[],"title_canon_sha256":"ff1c37e7fd77a6886a358fe9c959eddb33db0f87dcd6fb4cdb4b89e9fcbcc451","abstract_canon_sha256":"311c1c87ca24dd8f7ade4c96085550cff69828150c366b90460bb0df727fc0cb"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T02:04:15.657516Z","signature_b64":"985V9xZpIkECtsHsDh9mgjcbyuN3UjTq6ulkpPbTR99xa7TUMHWEy4xSjRD3+tCjDOAOHNFgV+L49XdOq4qhDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"386e4b0acd73336b232853e2b467e197500928817d12731e511b5940f4820594","last_reissued_at":"2026-07-05T02:04:15.657104Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T02:04:15.657104Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Multi-task Retrieval for Knowledge-Intensive Tasks","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Barlas O\\u{g}uz, Fabio Petroni, Gargi Ghosh, Jean Maillard, Veselin Stoyanov, Vladimir Karpukhin, Wen-tau Yih","submitted_at":"2021-01-01T00:16:34Z","abstract_excerpt":"Retrieving relevant contexts from a large corpus is a crucial step for tasks such as open-domain question answering and fact checking. Although neural retrieval outperforms traditional methods like tf-idf and BM25, its performance degrades considerably when applied to out-of-domain data.\n  Driven by the question of whether a neural retrieval model can be universal and perform robustly on a wide variety of problems, we propose a multi-task trained model. Our approach not only outperforms previous methods in the few-shot setting, but also rivals specialised neural retrievers, even when in-domain"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2101.00117","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/2101.00117/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":"2101.00117","created_at":"2026-07-05T02:04:15.657151+00:00"},{"alias_kind":"arxiv_version","alias_value":"2101.00117v1","created_at":"2026-07-05T02:04:15.657151+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2101.00117","created_at":"2026-07-05T02:04:15.657151+00:00"},{"alias_kind":"pith_short_12","alias_value":"HBXEWCWNOMZW","created_at":"2026-07-05T02:04:15.657151+00:00"},{"alias_kind":"pith_short_16","alias_value":"HBXEWCWNOMZWWIZI","created_at":"2026-07-05T02:04:15.657151+00:00"},{"alias_kind":"pith_short_8","alias_value":"HBXEWCWN","created_at":"2026-07-05T02:04:15.657151+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2501.04652","citing_title":"Multi-task retriever fine-tuning for domain-specific and efficient RAG","ref_index":15,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/HBXEWCWNOMZWWIZIKPRLIZ7BS5","json":"https://pith.science/pith/HBXEWCWNOMZWWIZIKPRLIZ7BS5.json","graph_json":"https://pith.science/api/pith-number/HBXEWCWNOMZWWIZIKPRLIZ7BS5/graph.json","events_json":"https://pith.science/api/pith-number/HBXEWCWNOMZWWIZIKPRLIZ7BS5/events.json","paper":"https://pith.science/paper/HBXEWCWN"},"agent_actions":{"view_html":"https://pith.science/pith/HBXEWCWNOMZWWIZIKPRLIZ7BS5","download_json":"https://pith.science/pith/HBXEWCWNOMZWWIZIKPRLIZ7BS5.json","view_paper":"https://pith.science/paper/HBXEWCWN","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2101.00117&json=true","fetch_graph":"https://pith.science/api/pith-number/HBXEWCWNOMZWWIZIKPRLIZ7BS5/graph.json","fetch_events":"https://pith.science/api/pith-number/HBXEWCWNOMZWWIZIKPRLIZ7BS5/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/HBXEWCWNOMZWWIZIKPRLIZ7BS5/action/timestamp_anchor","attest_storage":"https://pith.science/pith/HBXEWCWNOMZWWIZIKPRLIZ7BS5/action/storage_attestation","attest_author":"https://pith.science/pith/HBXEWCWNOMZWWIZIKPRLIZ7BS5/action/author_attestation","sign_citation":"https://pith.science/pith/HBXEWCWNOMZWWIZIKPRLIZ7BS5/action/citation_signature","submit_replication":"https://pith.science/pith/HBXEWCWNOMZWWIZIKPRLIZ7BS5/action/replication_record"}},"created_at":"2026-07-05T02:04:15.657151+00:00","updated_at":"2026-07-05T02:04:15.657151+00:00"}