{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2020:EUSPMHCIKN5OVH65FVY2QOIHZO","short_pith_number":"pith:EUSPMHCI","schema_version":"1.0","canonical_sha256":"2524f61c48537aea9fdd2d71a83907cb9babb1b5fe1c1a9185ea141d4bffcc65","source":{"kind":"arxiv","id":"2007.04612","version":3},"attestation_state":"computed","paper":{"title":"Concept Bottleneck Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Been Kim, Emma Pierson, Pang Wei Koh, Percy Liang, Stephen Mussmann, Thao Nguyen, Yew Siang Tang","submitted_at":"2020-07-09T07:47:28Z","abstract_excerpt":"We seek to learn models that we can interact with using high-level concepts: if the model did not think there was a bone spur in the x-ray, would it still predict severe arthritis? State-of-the-art models today do not typically support the manipulation of concepts like \"the existence of bone spurs\", as they are trained end-to-end to go directly from raw input (e.g., pixels) to output (e.g., arthritis severity). We revisit the classic idea of first predicting concepts that are provided at training time, and then using these concepts to predict the label. By construction, we can intervene on 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":"2007.04612","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2020-07-09T07:47:28Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"36e1ce81dbeaace97f2bd7d37c849ec36001eb335090baca83fe1d8e359e3e34","abstract_canon_sha256":"9b2b87a8b22a0a918d4215d0f088847507d33d8f4f059c8982724e457ae9e350"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T02:02:54.529705Z","signature_b64":"DVZegwwb8bObiLcL/zvL5CSAJ+EHaANEPF3sLVxwbyRf3pcJzJGubQSVvihxh2x3f7Zqku9pFwhiQ4c3VQWmDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"2524f61c48537aea9fdd2d71a83907cb9babb1b5fe1c1a9185ea141d4bffcc65","last_reissued_at":"2026-07-05T02:02:54.529260Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T02:02:54.529260Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Concept Bottleneck Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Been Kim, Emma Pierson, Pang Wei Koh, Percy Liang, Stephen Mussmann, Thao Nguyen, Yew Siang Tang","submitted_at":"2020-07-09T07:47:28Z","abstract_excerpt":"We seek to learn models that we can interact with using high-level concepts: if the model did not think there was a bone spur in the x-ray, would it still predict severe arthritis? State-of-the-art models today do not typically support the manipulation of concepts like \"the existence of bone spurs\", as they are trained end-to-end to go directly from raw input (e.g., pixels) to output (e.g., arthritis severity). We revisit the classic idea of first predicting concepts that are provided at training time, and then using these concepts to predict the label. By construction, we can intervene on the"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2007.04612","kind":"arxiv","version":3},"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/2007.04612/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":"2007.04612","created_at":"2026-07-05T02:02:54.529322+00:00"},{"alias_kind":"arxiv_version","alias_value":"2007.04612v3","created_at":"2026-07-05T02:02:54.529322+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2007.04612","created_at":"2026-07-05T02:02:54.529322+00:00"},{"alias_kind":"pith_short_12","alias_value":"EUSPMHCIKN5O","created_at":"2026-07-05T02:02:54.529322+00:00"},{"alias_kind":"pith_short_16","alias_value":"EUSPMHCIKN5OVH65","created_at":"2026-07-05T02:02:54.529322+00:00"},{"alias_kind":"pith_short_8","alias_value":"EUSPMHCI","created_at":"2026-07-05T02:02:54.529322+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":10,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2607.05175","citing_title":"Platonic Projection Structures: Operator-Induced Observability in Representation Learning","ref_index":33,"is_internal_anchor":true},{"citing_arxiv_id":"2607.01648","citing_title":"Boosting Ultrasound Image Classification via Attribute-Guided Dual-Branch Framework","ref_index":22,"is_internal_anchor":false},{"citing_arxiv_id":"2606.08032","citing_title":"Variational Proximal Policy Optimization","ref_index":62,"is_internal_anchor":false},{"citing_arxiv_id":"2606.04326","citing_title":"Measuring What Matters: Synthetic Benchmarks for Concept Bottleneck Models","ref_index":49,"is_internal_anchor":false},{"citing_arxiv_id":"2606.00629","citing_title":"Quality Audio Prototyping: a prototype system for unified sound retrieval and procedural generation","ref_index":30,"is_internal_anchor":false},{"citing_arxiv_id":"2602.05126","citing_title":"CLEAR-HPV: Interpretable concept discovery for human-papillomavirus-associated morphology in whole-slide histology","ref_index":22,"is_internal_anchor":false},{"citing_arxiv_id":"2602.05126","citing_title":"CLEAR-HPV: Interpretable concept discovery for human-papillomavirus-associated morphology in whole-slide histology","ref_index":22,"is_internal_anchor":false},{"citing_arxiv_id":"2605.21295","citing_title":"TimeSRL: Generalizable Time-Series Behavioral Modeling via Semantic RL-Tuned LLMs -- A Case Study in Mental Health","ref_index":26,"is_internal_anchor":false},{"citing_arxiv_id":"2602.24176","citing_title":"Beyond Explainable AI (XAI): An Overdue Paradigm Shift and Post-XAI Research Directions","ref_index":272,"is_internal_anchor":false},{"citing_arxiv_id":"2605.06640","citing_title":"Concept-Based Abductive and Contrastive Explanations for Behaviors of Vision Models","ref_index":68,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/EUSPMHCIKN5OVH65FVY2QOIHZO","json":"https://pith.science/pith/EUSPMHCIKN5OVH65FVY2QOIHZO.json","graph_json":"https://pith.science/api/pith-number/EUSPMHCIKN5OVH65FVY2QOIHZO/graph.json","events_json":"https://pith.science/api/pith-number/EUSPMHCIKN5OVH65FVY2QOIHZO/events.json","paper":"https://pith.science/paper/EUSPMHCI"},"agent_actions":{"view_html":"https://pith.science/pith/EUSPMHCIKN5OVH65FVY2QOIHZO","download_json":"https://pith.science/pith/EUSPMHCIKN5OVH65FVY2QOIHZO.json","view_paper":"https://pith.science/paper/EUSPMHCI","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2007.04612&json=true","fetch_graph":"https://pith.science/api/pith-number/EUSPMHCIKN5OVH65FVY2QOIHZO/graph.json","fetch_events":"https://pith.science/api/pith-number/EUSPMHCIKN5OVH65FVY2QOIHZO/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/EUSPMHCIKN5OVH65FVY2QOIHZO/action/timestamp_anchor","attest_storage":"https://pith.science/pith/EUSPMHCIKN5OVH65FVY2QOIHZO/action/storage_attestation","attest_author":"https://pith.science/pith/EUSPMHCIKN5OVH65FVY2QOIHZO/action/author_attestation","sign_citation":"https://pith.science/pith/EUSPMHCIKN5OVH65FVY2QOIHZO/action/citation_signature","submit_replication":"https://pith.science/pith/EUSPMHCIKN5OVH65FVY2QOIHZO/action/replication_record"}},"created_at":"2026-07-05T02:02:54.529322+00:00","updated_at":"2026-07-05T02:02:54.529322+00:00"}