{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:C2QROUHFNRLLTCRKYCONI3FJOM","short_pith_number":"pith:C2QROUHF","schema_version":"1.0","canonical_sha256":"16a11750e56c56b98a2ac09cd46ca973187cad27afc97df1a4871a9cf2d7fdd7","source":{"kind":"arxiv","id":"2501.09333","version":2},"attestation_state":"computed","paper":{"title":"Prompt-CAM: Making Vision Transformers Interpretable for Fine-Grained Analysis","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Anuj Karpatne, Arpita Chowdhury, Charles V. Stewart, Daniel Rubenstein, Dipanjyoti Paul, Elizabeth G. Campolongo, Jianyang Gu, Kazi Sajeed Mehrab, Tanya Berger-Wolf, Wei-Lun Chao, Yu Su, Zheda Mai, Ziheng Zhang","submitted_at":"2025-01-16T07:07:41Z","abstract_excerpt":"We present a simple approach to make pre-trained Vision Transformers (ViTs) interpretable for fine-grained analysis, aiming to identify and localize the traits that distinguish visually similar categories, such as bird species. Pre-trained ViTs, such as DINO, have demonstrated remarkable capabilities in extracting localized, discriminative features. However, saliency maps like Grad-CAM often fail to identify these traits, producing blurred, coarse heatmaps that highlight entire objects instead. We propose a novel approach, Prompt Class Attention Map (Prompt-CAM), to address this limitation. Pr"},"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":"2501.09333","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CV","submitted_at":"2025-01-16T07:07:41Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"0a1dace2392bc816a4abbf62d205838f6447a6a29fdabb11998918285f5cae5e","abstract_canon_sha256":"4565df2817140a34c456f7c86f1e218737e906be0ea68d25dd7c173bee32168d"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:45:38.870460Z","signature_b64":"E32w7Bzo3xrYrMc9QGBcT/U4ngYOZTPxg6NTIaD1mazbEgFop/K0/KcnYjsRz4nbBfctQhKqAPSHHlvXoyuqCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"16a11750e56c56b98a2ac09cd46ca973187cad27afc97df1a4871a9cf2d7fdd7","last_reissued_at":"2026-07-05T10:45:38.869956Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:45:38.869956Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Prompt-CAM: Making Vision Transformers Interpretable for Fine-Grained Analysis","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Anuj Karpatne, Arpita Chowdhury, Charles V. Stewart, Daniel Rubenstein, Dipanjyoti Paul, Elizabeth G. Campolongo, Jianyang Gu, Kazi Sajeed Mehrab, Tanya Berger-Wolf, Wei-Lun Chao, Yu Su, Zheda Mai, Ziheng Zhang","submitted_at":"2025-01-16T07:07:41Z","abstract_excerpt":"We present a simple approach to make pre-trained Vision Transformers (ViTs) interpretable for fine-grained analysis, aiming to identify and localize the traits that distinguish visually similar categories, such as bird species. Pre-trained ViTs, such as DINO, have demonstrated remarkable capabilities in extracting localized, discriminative features. However, saliency maps like Grad-CAM often fail to identify these traits, producing blurred, coarse heatmaps that highlight entire objects instead. We propose a novel approach, Prompt Class Attention Map (Prompt-CAM), to address this limitation. Pr"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2501.09333","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/2501.09333/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":"2501.09333","created_at":"2026-07-05T10:45:38.870014+00:00"},{"alias_kind":"arxiv_version","alias_value":"2501.09333v2","created_at":"2026-07-05T10:45:38.870014+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2501.09333","created_at":"2026-07-05T10:45:38.870014+00:00"},{"alias_kind":"pith_short_12","alias_value":"C2QROUHFNRLL","created_at":"2026-07-05T10:45:38.870014+00:00"},{"alias_kind":"pith_short_16","alias_value":"C2QROUHFNRLLTCRK","created_at":"2026-07-05T10:45:38.870014+00:00"},{"alias_kind":"pith_short_8","alias_value":"C2QROUHF","created_at":"2026-07-05T10:45:38.870014+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.18885","citing_title":"LARE: Low-Attention Region Encoding for Text-Image Retrieval","ref_index":60,"is_internal_anchor":false},{"citing_arxiv_id":"2506.09082","citing_title":"AVA-Bench: Atomic Visual Ability Benchmark for Vision Foundation Models","ref_index":14,"is_internal_anchor":false},{"citing_arxiv_id":"2507.14787","citing_title":"FOCUS: Fused Observation of Channels for Unveiling Spectra","ref_index":8,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/C2QROUHFNRLLTCRKYCONI3FJOM","json":"https://pith.science/pith/C2QROUHFNRLLTCRKYCONI3FJOM.json","graph_json":"https://pith.science/api/pith-number/C2QROUHFNRLLTCRKYCONI3FJOM/graph.json","events_json":"https://pith.science/api/pith-number/C2QROUHFNRLLTCRKYCONI3FJOM/events.json","paper":"https://pith.science/paper/C2QROUHF"},"agent_actions":{"view_html":"https://pith.science/pith/C2QROUHFNRLLTCRKYCONI3FJOM","download_json":"https://pith.science/pith/C2QROUHFNRLLTCRKYCONI3FJOM.json","view_paper":"https://pith.science/paper/C2QROUHF","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2501.09333&json=true","fetch_graph":"https://pith.science/api/pith-number/C2QROUHFNRLLTCRKYCONI3FJOM/graph.json","fetch_events":"https://pith.science/api/pith-number/C2QROUHFNRLLTCRKYCONI3FJOM/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/C2QROUHFNRLLTCRKYCONI3FJOM/action/timestamp_anchor","attest_storage":"https://pith.science/pith/C2QROUHFNRLLTCRKYCONI3FJOM/action/storage_attestation","attest_author":"https://pith.science/pith/C2QROUHFNRLLTCRKYCONI3FJOM/action/author_attestation","sign_citation":"https://pith.science/pith/C2QROUHFNRLLTCRKYCONI3FJOM/action/citation_signature","submit_replication":"https://pith.science/pith/C2QROUHFNRLLTCRKYCONI3FJOM/action/replication_record"}},"created_at":"2026-07-05T10:45:38.870014+00:00","updated_at":"2026-07-05T10:45:38.870014+00:00"}