{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:PKFEJU7YF4KEDMUT3DIVED2U57","short_pith_number":"pith:PKFEJU7Y","schema_version":"1.0","canonical_sha256":"7a8a44d3f82f1441b293d8d1520f54efd682852d6c4b1ea1099844631c7ac619","source":{"kind":"arxiv","id":"2508.14562","version":1},"attestation_state":"computed","paper":{"title":"Locality-aware Concept Bottleneck Model","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Byoung-Tak Zhang, Eungseo Kim, Hyundo Lee, Inwoo Hwang, Sanghack Lee, Sujin Jeon","submitted_at":"2025-08-20T09:33:48Z","abstract_excerpt":"Concept bottleneck models (CBMs) are inherently interpretable models that make predictions based on human-understandable visual cues, referred to as concepts. As obtaining dense concept annotations with human labeling is demanding and costly, recent approaches utilize foundation models to determine the concepts existing in the images. However, such label-free CBMs often fail to localize concepts in relevant regions, attending to visually unrelated regions when predicting concept presence. To this end, we propose a framework, coined Locality-aware Concept Bottleneck Model (LCBM), which utilizes"},"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":"2508.14562","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-08-20T09:33:48Z","cross_cats_sorted":[],"title_canon_sha256":"3eeb4fac93fb6a78ebfa68579413c48f3141f05f8f3d50210f2f936c47b6c4da","abstract_canon_sha256":"bf4f73cb7f108ab0bf1b77bd7bd17ab512fe656bc12d8c0e63cf6ed6d3326109"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:56:34.319737Z","signature_b64":"h37H4+60JxDBFjQmfC0EfUgngj1OaZRCgNyPh596KHj8miE+vN5Nzzy+EMa2wXuTmoRWTwEax1V35uISAb42Bg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7a8a44d3f82f1441b293d8d1520f54efd682852d6c4b1ea1099844631c7ac619","last_reissued_at":"2026-07-05T11:56:34.319225Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:56:34.319225Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Locality-aware Concept Bottleneck Model","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Byoung-Tak Zhang, Eungseo Kim, Hyundo Lee, Inwoo Hwang, Sanghack Lee, Sujin Jeon","submitted_at":"2025-08-20T09:33:48Z","abstract_excerpt":"Concept bottleneck models (CBMs) are inherently interpretable models that make predictions based on human-understandable visual cues, referred to as concepts. As obtaining dense concept annotations with human labeling is demanding and costly, recent approaches utilize foundation models to determine the concepts existing in the images. However, such label-free CBMs often fail to localize concepts in relevant regions, attending to visually unrelated regions when predicting concept presence. To this end, we propose a framework, coined Locality-aware Concept Bottleneck Model (LCBM), which utilizes"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2508.14562","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/2508.14562/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":"2508.14562","created_at":"2026-07-05T11:56:34.319285+00:00"},{"alias_kind":"arxiv_version","alias_value":"2508.14562v1","created_at":"2026-07-05T11:56:34.319285+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2508.14562","created_at":"2026-07-05T11:56:34.319285+00:00"},{"alias_kind":"pith_short_12","alias_value":"PKFEJU7YF4KE","created_at":"2026-07-05T11:56:34.319285+00:00"},{"alias_kind":"pith_short_16","alias_value":"PKFEJU7YF4KEDMUT","created_at":"2026-07-05T11:56:34.319285+00:00"},{"alias_kind":"pith_short_8","alias_value":"PKFEJU7Y","created_at":"2026-07-05T11:56:34.319285+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.04326","citing_title":"Measuring What Matters: Synthetic Benchmarks for Concept Bottleneck Models","ref_index":39,"is_internal_anchor":false},{"citing_arxiv_id":"2604.16076","citing_title":"Prototype-Grounded Concept Models for Verifiable Concept Alignment","ref_index":13,"is_internal_anchor":false},{"citing_arxiv_id":"2604.16076","citing_title":"Prototype-Grounded Concept Models for Verifiable Concept Alignment","ref_index":13,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/PKFEJU7YF4KEDMUT3DIVED2U57","json":"https://pith.science/pith/PKFEJU7YF4KEDMUT3DIVED2U57.json","graph_json":"https://pith.science/api/pith-number/PKFEJU7YF4KEDMUT3DIVED2U57/graph.json","events_json":"https://pith.science/api/pith-number/PKFEJU7YF4KEDMUT3DIVED2U57/events.json","paper":"https://pith.science/paper/PKFEJU7Y"},"agent_actions":{"view_html":"https://pith.science/pith/PKFEJU7YF4KEDMUT3DIVED2U57","download_json":"https://pith.science/pith/PKFEJU7YF4KEDMUT3DIVED2U57.json","view_paper":"https://pith.science/paper/PKFEJU7Y","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2508.14562&json=true","fetch_graph":"https://pith.science/api/pith-number/PKFEJU7YF4KEDMUT3DIVED2U57/graph.json","fetch_events":"https://pith.science/api/pith-number/PKFEJU7YF4KEDMUT3DIVED2U57/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/PKFEJU7YF4KEDMUT3DIVED2U57/action/timestamp_anchor","attest_storage":"https://pith.science/pith/PKFEJU7YF4KEDMUT3DIVED2U57/action/storage_attestation","attest_author":"https://pith.science/pith/PKFEJU7YF4KEDMUT3DIVED2U57/action/author_attestation","sign_citation":"https://pith.science/pith/PKFEJU7YF4KEDMUT3DIVED2U57/action/citation_signature","submit_replication":"https://pith.science/pith/PKFEJU7YF4KEDMUT3DIVED2U57/action/replication_record"}},"created_at":"2026-07-05T11:56:34.319285+00:00","updated_at":"2026-07-05T11:56:34.319285+00:00"}