{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2016:E42HXJJ7CNDPJMPCZ2MKBBD4MC","short_pith_number":"pith:E42HXJJ7","schema_version":"1.0","canonical_sha256":"27347ba53f1346f4b1e2ce98a0847c60919963d726dbf371f2cc1d9929708c74","source":{"kind":"arxiv","id":"1611.07450","version":2},"attestation_state":"computed","paper":{"title":"Grad-CAM: Why did you say that?","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CV","cs.LG"],"primary_cat":"stat.ML","authors_text":"Abhishek Das, Devi Parikh, Dhruv Batra, Michael Cogswell, Ramakrishna Vedantam, Ramprasaath R Selvaraju","submitted_at":"2016-11-22T18:34:36Z","abstract_excerpt":"We propose a technique for making Convolutional Neural Network (CNN)-based models more transparent by visualizing input regions that are 'important' for predictions -- or visual explanations. Our approach, called Gradient-weighted Class Activation Mapping (Grad-CAM), uses class-specific gradient information to localize important regions. These localizations are combined with existing pixel-space visualizations to create a novel high-resolution and class-discriminative visualization called Guided Grad-CAM. These methods help better understand CNN-based models, including image captioning and vis"},"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":"1611.07450","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"stat.ML","submitted_at":"2016-11-22T18:34:36Z","cross_cats_sorted":["cs.CV","cs.LG"],"title_canon_sha256":"8b61aec900b51264781ec89e540c881145dd3289a0c68327f79c6f5b91215c6e","abstract_canon_sha256":"f68fac44245227f940177da98b5e2989fd3a04e50b0b651798563e5350d1990b"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T00:52:07.953654Z","signature_b64":"mNSxvdFxggvWmgAib9XKfqn1/pPoDrbOhSuGN8vMW0vhvRklnVB5222/Iq8hZ9Tp4EBtsxp/fjQGpC0p7g2gAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"27347ba53f1346f4b1e2ce98a0847c60919963d726dbf371f2cc1d9929708c74","last_reissued_at":"2026-05-18T00:52:07.953091Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T00:52:07.953091Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Grad-CAM: Why did you say that?","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CV","cs.LG"],"primary_cat":"stat.ML","authors_text":"Abhishek Das, Devi Parikh, Dhruv Batra, Michael Cogswell, Ramakrishna Vedantam, Ramprasaath R Selvaraju","submitted_at":"2016-11-22T18:34:36Z","abstract_excerpt":"We propose a technique for making Convolutional Neural Network (CNN)-based models more transparent by visualizing input regions that are 'important' for predictions -- or visual explanations. Our approach, called Gradient-weighted Class Activation Mapping (Grad-CAM), uses class-specific gradient information to localize important regions. These localizations are combined with existing pixel-space visualizations to create a novel high-resolution and class-discriminative visualization called Guided Grad-CAM. These methods help better understand CNN-based models, including image captioning and vis"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1611.07450","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":""},"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":"1611.07450","created_at":"2026-05-18T00:52:07.953171+00:00"},{"alias_kind":"arxiv_version","alias_value":"1611.07450v2","created_at":"2026-05-18T00:52:07.953171+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1611.07450","created_at":"2026-05-18T00:52:07.953171+00:00"},{"alias_kind":"pith_short_12","alias_value":"E42HXJJ7CNDP","created_at":"2026-05-18T12:30:12.583610+00:00"},{"alias_kind":"pith_short_16","alias_value":"E42HXJJ7CNDPJMPC","created_at":"2026-05-18T12:30:12.583610+00:00"},{"alias_kind":"pith_short_8","alias_value":"E42HXJJ7","created_at":"2026-05-18T12:30:12.583610+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2601.11842","citing_title":"Template-free search for gravitational wave events using coincident anomaly detection","ref_index":27,"is_internal_anchor":true},{"citing_arxiv_id":"2604.23354","citing_title":"Explainable AI in Speaker Recognition -- Making Latent Representations Understandable","ref_index":32,"is_internal_anchor":false},{"citing_arxiv_id":"2605.00889","citing_title":"On the explainability of max-plus neural networks","ref_index":13,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/E42HXJJ7CNDPJMPCZ2MKBBD4MC","json":"https://pith.science/pith/E42HXJJ7CNDPJMPCZ2MKBBD4MC.json","graph_json":"https://pith.science/api/pith-number/E42HXJJ7CNDPJMPCZ2MKBBD4MC/graph.json","events_json":"https://pith.science/api/pith-number/E42HXJJ7CNDPJMPCZ2MKBBD4MC/events.json","paper":"https://pith.science/paper/E42HXJJ7"},"agent_actions":{"view_html":"https://pith.science/pith/E42HXJJ7CNDPJMPCZ2MKBBD4MC","download_json":"https://pith.science/pith/E42HXJJ7CNDPJMPCZ2MKBBD4MC.json","view_paper":"https://pith.science/paper/E42HXJJ7","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1611.07450&json=true","fetch_graph":"https://pith.science/api/pith-number/E42HXJJ7CNDPJMPCZ2MKBBD4MC/graph.json","fetch_events":"https://pith.science/api/pith-number/E42HXJJ7CNDPJMPCZ2MKBBD4MC/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/E42HXJJ7CNDPJMPCZ2MKBBD4MC/action/timestamp_anchor","attest_storage":"https://pith.science/pith/E42HXJJ7CNDPJMPCZ2MKBBD4MC/action/storage_attestation","attest_author":"https://pith.science/pith/E42HXJJ7CNDPJMPCZ2MKBBD4MC/action/author_attestation","sign_citation":"https://pith.science/pith/E42HXJJ7CNDPJMPCZ2MKBBD4MC/action/citation_signature","submit_replication":"https://pith.science/pith/E42HXJJ7CNDPJMPCZ2MKBBD4MC/action/replication_record"}},"created_at":"2026-05-18T00:52:07.953171+00:00","updated_at":"2026-05-18T00:52:07.953171+00:00"}