{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:EML7L7ZZZVARK4EJKPR47GNSDY","short_pith_number":"pith:EML7L7ZZ","schema_version":"1.0","canonical_sha256":"2317f5ff39cd4115708953e3cf99b21e38d42f7ae05dab3f9b1145e4a4bd4858","source":{"kind":"arxiv","id":"2401.06910","version":1},"attestation_state":"computed","paper":{"title":"InRanker: Distilled Rankers for Zero-shot Information Retrieval","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.IR","authors_text":"Konstantinos Papakostas, Roberto Lotufo, Rodrigo Nogueira, Thiago Laitz","submitted_at":"2024-01-12T21:52:42Z","abstract_excerpt":"Despite multi-billion parameter neural rankers being common components of state-of-the-art information retrieval pipelines, they are rarely used in production due to the enormous amount of compute required for inference. In this work, we propose a new method for distilling large rankers into their smaller versions focusing on out-of-domain effectiveness. We introduce InRanker, a version of monoT5 distilled from monoT5-3B with increased effectiveness on out-of-domain scenarios. Our key insight is to use language models and rerankers to generate as much as possible synthetic \"in-domain\" training"},"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":"2401.06910","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.IR","submitted_at":"2024-01-12T21:52:42Z","cross_cats_sorted":[],"title_canon_sha256":"bd84ff6e1571e2b2076e0b8b78abdbd136d84c2c05f234a3d1cc5063af58496a","abstract_canon_sha256":"fd09bf6151e4ed46602ab38edff257b371b9e30bfd715f6bde07945d69672287"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:33:28.022426Z","signature_b64":"5MMt2RWeHMVzNSBmmu3Lb4qnCD8JFDlRlp+BU5ZLt9JjFUaE3DV2AnwtDvzgpNCyvgAN5u7xhLKEUBdiMTxECQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"2317f5ff39cd4115708953e3cf99b21e38d42f7ae05dab3f9b1145e4a4bd4858","last_reissued_at":"2026-07-05T07:33:28.021983Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:33:28.021983Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"InRanker: Distilled Rankers for Zero-shot Information Retrieval","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.IR","authors_text":"Konstantinos Papakostas, Roberto Lotufo, Rodrigo Nogueira, Thiago Laitz","submitted_at":"2024-01-12T21:52:42Z","abstract_excerpt":"Despite multi-billion parameter neural rankers being common components of state-of-the-art information retrieval pipelines, they are rarely used in production due to the enormous amount of compute required for inference. In this work, we propose a new method for distilling large rankers into their smaller versions focusing on out-of-domain effectiveness. We introduce InRanker, a version of monoT5 distilled from monoT5-3B with increased effectiveness on out-of-domain scenarios. Our key insight is to use language models and rerankers to generate as much as possible synthetic \"in-domain\" training"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2401.06910","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/2401.06910/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":"2401.06910","created_at":"2026-07-05T07:33:28.022042+00:00"},{"alias_kind":"arxiv_version","alias_value":"2401.06910v1","created_at":"2026-07-05T07:33:28.022042+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2401.06910","created_at":"2026-07-05T07:33:28.022042+00:00"},{"alias_kind":"pith_short_12","alias_value":"EML7L7ZZZVAR","created_at":"2026-07-05T07:33:28.022042+00:00"},{"alias_kind":"pith_short_16","alias_value":"EML7L7ZZZVARK4EJ","created_at":"2026-07-05T07:33:28.022042+00:00"},{"alias_kind":"pith_short_8","alias_value":"EML7L7ZZ","created_at":"2026-07-05T07:33:28.022042+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2506.12494","citing_title":"FlexRAG: A Flexible and Comprehensive Framework for Retrieval-Augmented Generation","ref_index":24,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/EML7L7ZZZVARK4EJKPR47GNSDY","json":"https://pith.science/pith/EML7L7ZZZVARK4EJKPR47GNSDY.json","graph_json":"https://pith.science/api/pith-number/EML7L7ZZZVARK4EJKPR47GNSDY/graph.json","events_json":"https://pith.science/api/pith-number/EML7L7ZZZVARK4EJKPR47GNSDY/events.json","paper":"https://pith.science/paper/EML7L7ZZ"},"agent_actions":{"view_html":"https://pith.science/pith/EML7L7ZZZVARK4EJKPR47GNSDY","download_json":"https://pith.science/pith/EML7L7ZZZVARK4EJKPR47GNSDY.json","view_paper":"https://pith.science/paper/EML7L7ZZ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2401.06910&json=true","fetch_graph":"https://pith.science/api/pith-number/EML7L7ZZZVARK4EJKPR47GNSDY/graph.json","fetch_events":"https://pith.science/api/pith-number/EML7L7ZZZVARK4EJKPR47GNSDY/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/EML7L7ZZZVARK4EJKPR47GNSDY/action/timestamp_anchor","attest_storage":"https://pith.science/pith/EML7L7ZZZVARK4EJKPR47GNSDY/action/storage_attestation","attest_author":"https://pith.science/pith/EML7L7ZZZVARK4EJKPR47GNSDY/action/author_attestation","sign_citation":"https://pith.science/pith/EML7L7ZZZVARK4EJKPR47GNSDY/action/citation_signature","submit_replication":"https://pith.science/pith/EML7L7ZZZVARK4EJKPR47GNSDY/action/replication_record"}},"created_at":"2026-07-05T07:33:28.022042+00:00","updated_at":"2026-07-05T07:33:28.022042+00:00"}