{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:VNRQUYWCBXSBPZ5IAB6CGLWGR6","short_pith_number":"pith:VNRQUYWC","schema_version":"1.0","canonical_sha256":"ab630a62c20de417e7a8007c232ec68f8b8fd9e7d808c3cb2e3a6ba3c2f07106","source":{"kind":"arxiv","id":"2607.00870","version":1},"attestation_state":"computed","paper":{"title":"Dynamic Bidirectional Pattern Memory: A Production-Scale Empirical Characterisation of Inference-Time Gating in Clinical NLP","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Ali H. Lazem, William Teahan","submitted_at":"2026-07-01T12:34:30Z","abstract_excerpt":"We study inference-time pattern-memory gating in a production-scale clinical natural language processing (NLP) pipeline. The pipeline pairs a generator (Llama-3.3 70B) proposing extractions with a verifier (MMed-Llama-3.1 70B) accepting or rejecting them, over 167,034 PMC-Patients narratives, and adds a lightweight memory that learns at deployment which extractions to filter, so the verifier need not re-examine candidates already seen to fail. We report four findings. First, learning filtering rules directly from the verifier's rejections failed at full scale: the relation-extraction filter st"},"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":"2607.00870","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.CL","submitted_at":"2026-07-01T12:34:30Z","cross_cats_sorted":[],"title_canon_sha256":"1abf7e1340ea3c451ecc766e384006a7300ca36000412cec4fd9e386b0b08df2","abstract_canon_sha256":"cc25656ae792c7e16afc29420678a28c3c46f42c786a5553cb2b618bb072cc5a"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-02T01:18:21.875515Z","signature_b64":"tyocUbO/OIOzwBxFF0puC8vkSHWE3rm//Kpw0EZDD9rRfKtRTlC1weD1PTITYW5MxVPfwtkgun1cQ0K7ihHRAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ab630a62c20de417e7a8007c232ec68f8b8fd9e7d808c3cb2e3a6ba3c2f07106","last_reissued_at":"2026-07-02T01:18:21.875199Z","signature_status":"signed_v1","first_computed_at":"2026-07-02T01:18:21.875199Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Dynamic Bidirectional Pattern Memory: A Production-Scale Empirical Characterisation of Inference-Time Gating in Clinical NLP","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Ali H. Lazem, William Teahan","submitted_at":"2026-07-01T12:34:30Z","abstract_excerpt":"We study inference-time pattern-memory gating in a production-scale clinical natural language processing (NLP) pipeline. The pipeline pairs a generator (Llama-3.3 70B) proposing extractions with a verifier (MMed-Llama-3.1 70B) accepting or rejecting them, over 167,034 PMC-Patients narratives, and adds a lightweight memory that learns at deployment which extractions to filter, so the verifier need not re-examine candidates already seen to fail. We report four findings. First, learning filtering rules directly from the verifier's rejections failed at full scale: the relation-extraction filter st"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.00870","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/2607.00870/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":"2607.00870","created_at":"2026-07-02T01:18:21.875244+00:00"},{"alias_kind":"arxiv_version","alias_value":"2607.00870v1","created_at":"2026-07-02T01:18:21.875244+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.00870","created_at":"2026-07-02T01:18:21.875244+00:00"},{"alias_kind":"pith_short_12","alias_value":"VNRQUYWCBXSB","created_at":"2026-07-02T01:18:21.875244+00:00"},{"alias_kind":"pith_short_16","alias_value":"VNRQUYWCBXSBPZ5I","created_at":"2026-07-02T01:18:21.875244+00:00"},{"alias_kind":"pith_short_8","alias_value":"VNRQUYWC","created_at":"2026-07-02T01:18:21.875244+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/VNRQUYWCBXSBPZ5IAB6CGLWGR6","json":"https://pith.science/pith/VNRQUYWCBXSBPZ5IAB6CGLWGR6.json","graph_json":"https://pith.science/api/pith-number/VNRQUYWCBXSBPZ5IAB6CGLWGR6/graph.json","events_json":"https://pith.science/api/pith-number/VNRQUYWCBXSBPZ5IAB6CGLWGR6/events.json","paper":"https://pith.science/paper/VNRQUYWC"},"agent_actions":{"view_html":"https://pith.science/pith/VNRQUYWCBXSBPZ5IAB6CGLWGR6","download_json":"https://pith.science/pith/VNRQUYWCBXSBPZ5IAB6CGLWGR6.json","view_paper":"https://pith.science/paper/VNRQUYWC","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2607.00870&json=true","fetch_graph":"https://pith.science/api/pith-number/VNRQUYWCBXSBPZ5IAB6CGLWGR6/graph.json","fetch_events":"https://pith.science/api/pith-number/VNRQUYWCBXSBPZ5IAB6CGLWGR6/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/VNRQUYWCBXSBPZ5IAB6CGLWGR6/action/timestamp_anchor","attest_storage":"https://pith.science/pith/VNRQUYWCBXSBPZ5IAB6CGLWGR6/action/storage_attestation","attest_author":"https://pith.science/pith/VNRQUYWCBXSBPZ5IAB6CGLWGR6/action/author_attestation","sign_citation":"https://pith.science/pith/VNRQUYWCBXSBPZ5IAB6CGLWGR6/action/citation_signature","submit_replication":"https://pith.science/pith/VNRQUYWCBXSBPZ5IAB6CGLWGR6/action/replication_record"}},"created_at":"2026-07-02T01:18:21.875244+00:00","updated_at":"2026-07-02T01:18:21.875244+00:00"}