{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:4XR33CSBW4CTKUEEXSULZYRRZK","short_pith_number":"pith:4XR33CSB","schema_version":"1.0","canonical_sha256":"e5e3bd8a41b705355084bca8bce231cab47f39433a555aac5448fc856a4958da","source":{"kind":"arxiv","id":"2505.03118","version":1},"attestation_state":"computed","paper":{"title":"Adaptive Thresholding for Multi-Label Classification via Global-Local Signal Fusion","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Dmytro Shamatrin","submitted_at":"2025-05-06T02:19:37Z","abstract_excerpt":"Multi-label classification (MLC) requires predicting multiple labels per sample, often under heavy class imbalance and noisy conditions. Traditional approaches apply fixed thresholds or treat labels independently, overlooking context and global rarity. We introduce an adaptive thresholding mechanism that fuses global (IDF-based) and local (KNN-based) signals to produce per-label, per-instance thresholds. Instead of applying these as hard cutoffs, we treat them as differentiable penalties in the loss, providing smooth supervision and better calibration. Our architecture is lightweight, interpre"},"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":"2505.03118","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-05-06T02:19:37Z","cross_cats_sorted":[],"title_canon_sha256":"4d383d3fc11d79aa533ddcd02fad0886232e0db413c99be6643df0180c588a01","abstract_canon_sha256":"17a2428949d7232ec765f9dd4a52bc9c6aaa1968f0e09d1cd768dba84b4792bb"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:58:52.896719Z","signature_b64":"L/B35CEn2oY0qJMnfZ22A7zG4OeYxuwWtHbxc8fS/0LmPG7lZFdywz76ZEB9qppFUQMmJDW9mc7LdjCcLaH3Cw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e5e3bd8a41b705355084bca8bce231cab47f39433a555aac5448fc856a4958da","last_reissued_at":"2026-07-05T10:58:52.896031Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:58:52.896031Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Adaptive Thresholding for Multi-Label Classification via Global-Local Signal Fusion","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Dmytro Shamatrin","submitted_at":"2025-05-06T02:19:37Z","abstract_excerpt":"Multi-label classification (MLC) requires predicting multiple labels per sample, often under heavy class imbalance and noisy conditions. Traditional approaches apply fixed thresholds or treat labels independently, overlooking context and global rarity. We introduce an adaptive thresholding mechanism that fuses global (IDF-based) and local (KNN-based) signals to produce per-label, per-instance thresholds. Instead of applying these as hard cutoffs, we treat them as differentiable penalties in the loss, providing smooth supervision and better calibration. Our architecture is lightweight, interpre"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.03118","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/2505.03118/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":"2505.03118","created_at":"2026-07-05T10:58:52.896228+00:00"},{"alias_kind":"arxiv_version","alias_value":"2505.03118v1","created_at":"2026-07-05T10:58:52.896228+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.03118","created_at":"2026-07-05T10:58:52.896228+00:00"},{"alias_kind":"pith_short_12","alias_value":"4XR33CSBW4CT","created_at":"2026-07-05T10:58:52.896228+00:00"},{"alias_kind":"pith_short_16","alias_value":"4XR33CSBW4CTKUEE","created_at":"2026-07-05T10:58:52.896228+00:00"},{"alias_kind":"pith_short_8","alias_value":"4XR33CSB","created_at":"2026-07-05T10:58:52.896228+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.16535","citing_title":"RAPT: Retrieval-Augmented Post-hoc Thresholding for Multi-Label Classification","ref_index":21,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/4XR33CSBW4CTKUEEXSULZYRRZK","json":"https://pith.science/pith/4XR33CSBW4CTKUEEXSULZYRRZK.json","graph_json":"https://pith.science/api/pith-number/4XR33CSBW4CTKUEEXSULZYRRZK/graph.json","events_json":"https://pith.science/api/pith-number/4XR33CSBW4CTKUEEXSULZYRRZK/events.json","paper":"https://pith.science/paper/4XR33CSB"},"agent_actions":{"view_html":"https://pith.science/pith/4XR33CSBW4CTKUEEXSULZYRRZK","download_json":"https://pith.science/pith/4XR33CSBW4CTKUEEXSULZYRRZK.json","view_paper":"https://pith.science/paper/4XR33CSB","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2505.03118&json=true","fetch_graph":"https://pith.science/api/pith-number/4XR33CSBW4CTKUEEXSULZYRRZK/graph.json","fetch_events":"https://pith.science/api/pith-number/4XR33CSBW4CTKUEEXSULZYRRZK/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/4XR33CSBW4CTKUEEXSULZYRRZK/action/timestamp_anchor","attest_storage":"https://pith.science/pith/4XR33CSBW4CTKUEEXSULZYRRZK/action/storage_attestation","attest_author":"https://pith.science/pith/4XR33CSBW4CTKUEEXSULZYRRZK/action/author_attestation","sign_citation":"https://pith.science/pith/4XR33CSBW4CTKUEEXSULZYRRZK/action/citation_signature","submit_replication":"https://pith.science/pith/4XR33CSBW4CTKUEEXSULZYRRZK/action/replication_record"}},"created_at":"2026-07-05T10:58:52.896228+00:00","updated_at":"2026-07-05T10:58:52.896228+00:00"}