{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:D7YNAPCN7ACJCWSJVAHPWIITKR","short_pith_number":"pith:D7YNAPCN","schema_version":"1.0","canonical_sha256":"1ff0d03c4df804915a49a80efb2113546a891d19235827899572bfee523ec2da","source":{"kind":"arxiv","id":"2412.10313","version":1},"attestation_state":"computed","paper":{"title":"MST-R: Multi-Stage Tuning for Retrieval Systems and Metric Evaluation","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.IR","authors_text":"Karan Dhingra, Maneesh Singh, Yash Malviya","submitted_at":"2024-12-13T17:53:29Z","abstract_excerpt":"Regulatory documents are rich in nuanced terminology and specialized semantics. FRAG systems: Frozen retrieval-augmented generators utilizing pre-trained (or, frozen) components face consequent challenges with both retriever and answering performance. We present a system that adapts the retriever performance to the target domain using a multi-stage tuning (MST) strategy. Our retrieval approach, called MST-R (a) first fine-tunes encoders used in vector stores using hard negative mining, (b) then uses a hybrid retriever, combining sparse and dense retrievers using reciprocal rank fusion, and the"},"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":"2412.10313","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.IR","submitted_at":"2024-12-13T17:53:29Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"53c8546547df34f732742982edec7388a3b8e93e69b3c948218bbe5d7f0d6090","abstract_canon_sha256":"b867313e7b6389f3759d9918d8492cc40474bd116786bbea274cb5ee9d313fd1"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:48:56.245019Z","signature_b64":"ZESJDBj/GCckWuPi88jXxu1YPddRSeqkIHRHaIRMT7fYuf6qXtBb0MYpT5P5XICZNz+WgrGZi6U7ydS+lqOzCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"1ff0d03c4df804915a49a80efb2113546a891d19235827899572bfee523ec2da","last_reissued_at":"2026-07-05T09:48:56.244497Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:48:56.244497Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"MST-R: Multi-Stage Tuning for Retrieval Systems and Metric Evaluation","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.IR","authors_text":"Karan Dhingra, Maneesh Singh, Yash Malviya","submitted_at":"2024-12-13T17:53:29Z","abstract_excerpt":"Regulatory documents are rich in nuanced terminology and specialized semantics. FRAG systems: Frozen retrieval-augmented generators utilizing pre-trained (or, frozen) components face consequent challenges with both retriever and answering performance. We present a system that adapts the retriever performance to the target domain using a multi-stage tuning (MST) strategy. Our retrieval approach, called MST-R (a) first fine-tunes encoders used in vector stores using hard negative mining, (b) then uses a hybrid retriever, combining sparse and dense retrievers using reciprocal rank fusion, and the"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2412.10313","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/2412.10313/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":"2412.10313","created_at":"2026-07-05T09:48:56.244562+00:00"},{"alias_kind":"arxiv_version","alias_value":"2412.10313v1","created_at":"2026-07-05T09:48:56.244562+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2412.10313","created_at":"2026-07-05T09:48:56.244562+00:00"},{"alias_kind":"pith_short_12","alias_value":"D7YNAPCN7ACJ","created_at":"2026-07-05T09:48:56.244562+00:00"},{"alias_kind":"pith_short_16","alias_value":"D7YNAPCN7ACJCWSJ","created_at":"2026-07-05T09:48:56.244562+00:00"},{"alias_kind":"pith_short_8","alias_value":"D7YNAPCN","created_at":"2026-07-05T09:48:56.244562+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/D7YNAPCN7ACJCWSJVAHPWIITKR","json":"https://pith.science/pith/D7YNAPCN7ACJCWSJVAHPWIITKR.json","graph_json":"https://pith.science/api/pith-number/D7YNAPCN7ACJCWSJVAHPWIITKR/graph.json","events_json":"https://pith.science/api/pith-number/D7YNAPCN7ACJCWSJVAHPWIITKR/events.json","paper":"https://pith.science/paper/D7YNAPCN"},"agent_actions":{"view_html":"https://pith.science/pith/D7YNAPCN7ACJCWSJVAHPWIITKR","download_json":"https://pith.science/pith/D7YNAPCN7ACJCWSJVAHPWIITKR.json","view_paper":"https://pith.science/paper/D7YNAPCN","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2412.10313&json=true","fetch_graph":"https://pith.science/api/pith-number/D7YNAPCN7ACJCWSJVAHPWIITKR/graph.json","fetch_events":"https://pith.science/api/pith-number/D7YNAPCN7ACJCWSJVAHPWIITKR/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/D7YNAPCN7ACJCWSJVAHPWIITKR/action/timestamp_anchor","attest_storage":"https://pith.science/pith/D7YNAPCN7ACJCWSJVAHPWIITKR/action/storage_attestation","attest_author":"https://pith.science/pith/D7YNAPCN7ACJCWSJVAHPWIITKR/action/author_attestation","sign_citation":"https://pith.science/pith/D7YNAPCN7ACJCWSJVAHPWIITKR/action/citation_signature","submit_replication":"https://pith.science/pith/D7YNAPCN7ACJCWSJVAHPWIITKR/action/replication_record"}},"created_at":"2026-07-05T09:48:56.244562+00:00","updated_at":"2026-07-05T09:48:56.244562+00:00"}