{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:SVCGCNMGB35RJWWLQJXLGJGF5M","short_pith_number":"pith:SVCGCNMG","schema_version":"1.0","canonical_sha256":"95446135860efb14dacb826eb324c5eb2578767ddf9a4a91e814f93fcf152de6","source":{"kind":"arxiv","id":"2510.04180","version":2},"attestation_state":"computed","paper":{"title":"Spatially Grounded Concept-Based Image Classification","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Amit Rozner, Ethan Fetaya, Ofir Lindenbaum, Ran Eisenberg","submitted_at":"2025-10-05T12:48:43Z","abstract_excerpt":"Deep neural networks can achieve high accuracy while relying on evidence that is hard to inspect or misaligned with the intended task. Concept Bottleneck Models (CBMs) expose human-interpretable concepts, but most treat concepts as global attributes and do not show how localized evidence is aggregated into a decision. We propose \\textbf{SEG-MIL-CBM}, a spatially grounded CBM that decomposes each image into concept-guided regions and classifies it by attention-based aggregation of segment-level concept evidence. The same segment evidence terms form the prediction and the explanation, exposing w"},"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":"2510.04180","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-10-05T12:48:43Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"2269ecb741d8b641e1777fd57e38cdb1769e4acbc4bff9d16f3fc2973cf6b719","abstract_canon_sha256":"4641421d69e009a8c99bf55b4127513b3a9992d487292ed0f21056e08f45f337"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-23T01:12:47.581426Z","signature_b64":"PlsF6ChOUIWvCG9LksucC4rvRwD0qdIQsZ4WlBf1Yqr1Al4GrhZfJ8wzBNFTe8rWZrdT5u6DsGIyp+5izXvFDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"95446135860efb14dacb826eb324c5eb2578767ddf9a4a91e814f93fcf152de6","last_reissued_at":"2026-06-23T01:12:47.580816Z","signature_status":"signed_v1","first_computed_at":"2026-06-23T01:12:47.580816Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Spatially Grounded Concept-Based Image Classification","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Amit Rozner, Ethan Fetaya, Ofir Lindenbaum, Ran Eisenberg","submitted_at":"2025-10-05T12:48:43Z","abstract_excerpt":"Deep neural networks can achieve high accuracy while relying on evidence that is hard to inspect or misaligned with the intended task. Concept Bottleneck Models (CBMs) expose human-interpretable concepts, but most treat concepts as global attributes and do not show how localized evidence is aggregated into a decision. We propose \\textbf{SEG-MIL-CBM}, a spatially grounded CBM that decomposes each image into concept-guided regions and classifies it by attention-based aggregation of segment-level concept evidence. The same segment evidence terms form the prediction and the explanation, exposing w"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2510.04180","kind":"arxiv","version":2},"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/2510.04180/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":"2510.04180","created_at":"2026-06-23T01:12:47.580889+00:00"},{"alias_kind":"arxiv_version","alias_value":"2510.04180v2","created_at":"2026-06-23T01:12:47.580889+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2510.04180","created_at":"2026-06-23T01:12:47.580889+00:00"},{"alias_kind":"pith_short_12","alias_value":"SVCGCNMGB35R","created_at":"2026-06-23T01:12:47.580889+00:00"},{"alias_kind":"pith_short_16","alias_value":"SVCGCNMGB35RJWWL","created_at":"2026-06-23T01:12:47.580889+00:00"},{"alias_kind":"pith_short_8","alias_value":"SVCGCNMG","created_at":"2026-06-23T01:12:47.580889+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/SVCGCNMGB35RJWWLQJXLGJGF5M","json":"https://pith.science/pith/SVCGCNMGB35RJWWLQJXLGJGF5M.json","graph_json":"https://pith.science/api/pith-number/SVCGCNMGB35RJWWLQJXLGJGF5M/graph.json","events_json":"https://pith.science/api/pith-number/SVCGCNMGB35RJWWLQJXLGJGF5M/events.json","paper":"https://pith.science/paper/SVCGCNMG"},"agent_actions":{"view_html":"https://pith.science/pith/SVCGCNMGB35RJWWLQJXLGJGF5M","download_json":"https://pith.science/pith/SVCGCNMGB35RJWWLQJXLGJGF5M.json","view_paper":"https://pith.science/paper/SVCGCNMG","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2510.04180&json=true","fetch_graph":"https://pith.science/api/pith-number/SVCGCNMGB35RJWWLQJXLGJGF5M/graph.json","fetch_events":"https://pith.science/api/pith-number/SVCGCNMGB35RJWWLQJXLGJGF5M/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/SVCGCNMGB35RJWWLQJXLGJGF5M/action/timestamp_anchor","attest_storage":"https://pith.science/pith/SVCGCNMGB35RJWWLQJXLGJGF5M/action/storage_attestation","attest_author":"https://pith.science/pith/SVCGCNMGB35RJWWLQJXLGJGF5M/action/author_attestation","sign_citation":"https://pith.science/pith/SVCGCNMGB35RJWWLQJXLGJGF5M/action/citation_signature","submit_replication":"https://pith.science/pith/SVCGCNMGB35RJWWLQJXLGJGF5M/action/replication_record"}},"created_at":"2026-06-23T01:12:47.580889+00:00","updated_at":"2026-06-23T01:12:47.580889+00:00"}