{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:M27ZMKCCKPZK5N4O5PDTI7PDKT","short_pith_number":"pith:M27ZMKCC","schema_version":"1.0","canonical_sha256":"66bf96284253f2aeb78eebc7347de354f9671316bfd808ca686f62dee877ab7a","source":{"kind":"arxiv","id":"2505.21439","version":1},"attestation_state":"computed","paper":{"title":"Towards Better Instruction Following Retrieval Models","license":"http://creativecommons.org/publicdomain/zero/1.0/","headline":"","cross_cats":["cs.IR"],"primary_cat":"cs.CL","authors_text":"Aaron Trinh, Bo Dai, Chao Zhang, Hanjun Dai, Haotian Sun, Rushi Qiang, Yuchen Zhuang","submitted_at":"2025-05-27T17:14:37Z","abstract_excerpt":"Modern information retrieval (IR) models, trained exclusively on standard <query, passage> pairs, struggle to effectively interpret and follow explicit user instructions. We introduce InF-IR, a large-scale, high-quality training corpus tailored for enhancing retrieval models in Instruction-Following IR. InF-IR expands traditional training pairs into over 38,000 expressive <instruction, query, passage> triplets as positive samples. In particular, for each positive triplet, we generate two additional hard negative examples by poisoning both instructions and queries, then rigorously validated by "},"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":"2505.21439","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/publicdomain/zero/1.0/","primary_cat":"cs.CL","submitted_at":"2025-05-27T17:14:37Z","cross_cats_sorted":["cs.IR"],"title_canon_sha256":"0e55598c44e4a2e33feec164f59cdd3952905e4920927efbdfae463a26e96892","abstract_canon_sha256":"99b48e628adb38411339ebeee7e03e77472d4de5b1b48ac7af3faa6aaf7e3d10"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:10:41.928058Z","signature_b64":"XPFGh/Z3sXJcSO5G0jjb7HPbwmjBQO7pVkqxrYXo2b6EUZZY3+XY9F4i+TAMU4KJtC8SW8UnH0d4bmZIk1NZBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"66bf96284253f2aeb78eebc7347de354f9671316bfd808ca686f62dee877ab7a","last_reissued_at":"2026-07-05T11:10:41.927556Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:10:41.927556Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Towards Better Instruction Following Retrieval Models","license":"http://creativecommons.org/publicdomain/zero/1.0/","headline":"","cross_cats":["cs.IR"],"primary_cat":"cs.CL","authors_text":"Aaron Trinh, Bo Dai, Chao Zhang, Hanjun Dai, Haotian Sun, Rushi Qiang, Yuchen Zhuang","submitted_at":"2025-05-27T17:14:37Z","abstract_excerpt":"Modern information retrieval (IR) models, trained exclusively on standard <query, passage> pairs, struggle to effectively interpret and follow explicit user instructions. We introduce InF-IR, a large-scale, high-quality training corpus tailored for enhancing retrieval models in Instruction-Following IR. InF-IR expands traditional training pairs into over 38,000 expressive <instruction, query, passage> triplets as positive samples. In particular, for each positive triplet, we generate two additional hard negative examples by poisoning both instructions and queries, then rigorously validated by "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.21439","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/2505.21439/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":"2505.21439","created_at":"2026-07-05T11:10:41.927616+00:00"},{"alias_kind":"arxiv_version","alias_value":"2505.21439v1","created_at":"2026-07-05T11:10:41.927616+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.21439","created_at":"2026-07-05T11:10:41.927616+00:00"},{"alias_kind":"pith_short_12","alias_value":"M27ZMKCCKPZK","created_at":"2026-07-05T11:10:41.927616+00:00"},{"alias_kind":"pith_short_16","alias_value":"M27ZMKCCKPZK5N4O","created_at":"2026-07-05T11:10:41.927616+00:00"},{"alias_kind":"pith_short_8","alias_value":"M27ZMKCC","created_at":"2026-07-05T11:10:41.927616+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/M27ZMKCCKPZK5N4O5PDTI7PDKT","json":"https://pith.science/pith/M27ZMKCCKPZK5N4O5PDTI7PDKT.json","graph_json":"https://pith.science/api/pith-number/M27ZMKCCKPZK5N4O5PDTI7PDKT/graph.json","events_json":"https://pith.science/api/pith-number/M27ZMKCCKPZK5N4O5PDTI7PDKT/events.json","paper":"https://pith.science/paper/M27ZMKCC"},"agent_actions":{"view_html":"https://pith.science/pith/M27ZMKCCKPZK5N4O5PDTI7PDKT","download_json":"https://pith.science/pith/M27ZMKCCKPZK5N4O5PDTI7PDKT.json","view_paper":"https://pith.science/paper/M27ZMKCC","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2505.21439&json=true","fetch_graph":"https://pith.science/api/pith-number/M27ZMKCCKPZK5N4O5PDTI7PDKT/graph.json","fetch_events":"https://pith.science/api/pith-number/M27ZMKCCKPZK5N4O5PDTI7PDKT/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/M27ZMKCCKPZK5N4O5PDTI7PDKT/action/timestamp_anchor","attest_storage":"https://pith.science/pith/M27ZMKCCKPZK5N4O5PDTI7PDKT/action/storage_attestation","attest_author":"https://pith.science/pith/M27ZMKCCKPZK5N4O5PDTI7PDKT/action/author_attestation","sign_citation":"https://pith.science/pith/M27ZMKCCKPZK5N4O5PDTI7PDKT/action/citation_signature","submit_replication":"https://pith.science/pith/M27ZMKCCKPZK5N4O5PDTI7PDKT/action/replication_record"}},"created_at":"2026-07-05T11:10:41.927616+00:00","updated_at":"2026-07-05T11:10:41.927616+00:00"}