{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:SELCHYQEY23D3HJQMYSKVQODJW","short_pith_number":"pith:SELCHYQE","schema_version":"1.0","canonical_sha256":"911623e204c6b63d9d306624aac1c34d9f7c6a0b38ddfd61f4a1670af22ac05f","source":{"kind":"arxiv","id":"2211.09260","version":2},"attestation_state":"computed","paper":{"title":"Task-aware Retrieval with Instructions","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Akari Asai, Gautier Izacard, Hannaneh Hajishirzi, Patrick Lewis, Sebastian Riedel, Timo Schick, Wen-tau Yih, Xilun Chen","submitted_at":"2022-11-16T23:13:22Z","abstract_excerpt":"We study the problem of retrieval with instructions, where users of a retrieval system explicitly describe their intent along with their queries. We aim to develop a general-purpose task-aware retrieval system using multi-task instruction tuning, which can follow human-written instructions to find the best documents for a given query. We introduce the first large-scale collection of approximately 40 retrieval datasets with instructions, BERRI, and present TART, a multi-task retrieval system trained on BERRI with instructions. TART shows strong capabilities to adapt to a new retrieval task via "},"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":"2211.09260","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2022-11-16T23:13:22Z","cross_cats_sorted":[],"title_canon_sha256":"3e39590e13d822161c19bd59c188e060f56cbfef9c805f4335fe7823864556b1","abstract_canon_sha256":"89670079e3f63a3323760bb89d9a24815fe2b1a151993d31cd49e79375f333f2"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:26:57.218245Z","signature_b64":"Jrl/Vt4xRyCEGyPeQp+5ivcZjSMnGXyiln0SVc0UzxubJvvwmOs+EmRCRFvQj4ROAKnv7d8Mq4di5JkQzAZuCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"911623e204c6b63d9d306624aac1c34d9f7c6a0b38ddfd61f4a1670af22ac05f","last_reissued_at":"2026-07-05T05:26:57.217815Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:26:57.217815Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Task-aware Retrieval with Instructions","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Akari Asai, Gautier Izacard, Hannaneh Hajishirzi, Patrick Lewis, Sebastian Riedel, Timo Schick, Wen-tau Yih, Xilun Chen","submitted_at":"2022-11-16T23:13:22Z","abstract_excerpt":"We study the problem of retrieval with instructions, where users of a retrieval system explicitly describe their intent along with their queries. We aim to develop a general-purpose task-aware retrieval system using multi-task instruction tuning, which can follow human-written instructions to find the best documents for a given query. We introduce the first large-scale collection of approximately 40 retrieval datasets with instructions, BERRI, and present TART, a multi-task retrieval system trained on BERRI with instructions. TART shows strong capabilities to adapt to a new retrieval task via "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2211.09260","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/2211.09260/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":"2211.09260","created_at":"2026-07-05T05:26:57.217869+00:00"},{"alias_kind":"arxiv_version","alias_value":"2211.09260v2","created_at":"2026-07-05T05:26:57.217869+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2211.09260","created_at":"2026-07-05T05:26:57.217869+00:00"},{"alias_kind":"pith_short_12","alias_value":"SELCHYQEY23D","created_at":"2026-07-05T05:26:57.217869+00:00"},{"alias_kind":"pith_short_16","alias_value":"SELCHYQEY23D3HJQ","created_at":"2026-07-05T05:26:57.217869+00:00"},{"alias_kind":"pith_short_8","alias_value":"SELCHYQE","created_at":"2026-07-05T05:26:57.217869+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":6,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.02411","citing_title":"FitText: Evolving Agent Tool Ecologies via Memetic Retrieval","ref_index":3,"is_internal_anchor":false},{"citing_arxiv_id":"2410.05160","citing_title":"VLM2Vec: Training Vision-Language Models for Massive Multimodal Embedding Tasks","ref_index":3,"is_internal_anchor":false},{"citing_arxiv_id":"2405.17428","citing_title":"NV-Embed: Improved Techniques for Training LLMs as Generalist Embedding Models","ref_index":134,"is_internal_anchor":false},{"citing_arxiv_id":"2309.07597","citing_title":"C-Pack: Packed Resources For General Chinese Embeddings","ref_index":9,"is_internal_anchor":false},{"citing_arxiv_id":"2310.11511","citing_title":"Self-RAG: Learning to Retrieve, Generate, and Critique through Self-Reflection","ref_index":101,"is_internal_anchor":false},{"citing_arxiv_id":"2605.02411","citing_title":"FitText: Evolving Agent Tool Ecologies via Memetic Retrieval","ref_index":2,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/SELCHYQEY23D3HJQMYSKVQODJW","json":"https://pith.science/pith/SELCHYQEY23D3HJQMYSKVQODJW.json","graph_json":"https://pith.science/api/pith-number/SELCHYQEY23D3HJQMYSKVQODJW/graph.json","events_json":"https://pith.science/api/pith-number/SELCHYQEY23D3HJQMYSKVQODJW/events.json","paper":"https://pith.science/paper/SELCHYQE"},"agent_actions":{"view_html":"https://pith.science/pith/SELCHYQEY23D3HJQMYSKVQODJW","download_json":"https://pith.science/pith/SELCHYQEY23D3HJQMYSKVQODJW.json","view_paper":"https://pith.science/paper/SELCHYQE","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2211.09260&json=true","fetch_graph":"https://pith.science/api/pith-number/SELCHYQEY23D3HJQMYSKVQODJW/graph.json","fetch_events":"https://pith.science/api/pith-number/SELCHYQEY23D3HJQMYSKVQODJW/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/SELCHYQEY23D3HJQMYSKVQODJW/action/timestamp_anchor","attest_storage":"https://pith.science/pith/SELCHYQEY23D3HJQMYSKVQODJW/action/storage_attestation","attest_author":"https://pith.science/pith/SELCHYQEY23D3HJQMYSKVQODJW/action/author_attestation","sign_citation":"https://pith.science/pith/SELCHYQEY23D3HJQMYSKVQODJW/action/citation_signature","submit_replication":"https://pith.science/pith/SELCHYQEY23D3HJQMYSKVQODJW/action/replication_record"}},"created_at":"2026-07-05T05:26:57.217869+00:00","updated_at":"2026-07-05T05:26:57.217869+00:00"}