{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:VB7HI67LI36EI6UIY5PZDE2X5C","short_pith_number":"pith:VB7HI67L","schema_version":"1.0","canonical_sha256":"a87e747beb46fc447a88c75f919357e891c52dff19cec0db7e8a3e9d5c31145e","source":{"kind":"arxiv","id":"2504.19414","version":1},"attestation_state":"computed","paper":{"title":"GMAR: Gradient-Driven Multi-Head Attention Rollout for Vision Transformer Interpretability","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Gangjae Jang, Haesol Park, Sehyeong Jo","submitted_at":"2025-04-28T01:58:39Z","abstract_excerpt":"The Vision Transformer (ViT) has made significant advancements in computer vision, utilizing self-attention mechanisms to achieve state-of-the-art performance across various tasks, including image classification, object detection, and segmentation. Its architectural flexibility and capabilities have made it a preferred choice among researchers and practitioners. However, the intricate multi-head attention mechanism of ViT presents significant challenges to interpretability, as the underlying prediction process remains opaque. A critical limitation arises from an observation commonly noted in t"},"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":"2504.19414","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2025-04-28T01:58:39Z","cross_cats_sorted":[],"title_canon_sha256":"476f5a6f7625dabc96f8df655cb093ba3e2eda89b81d52bef45507114661798c","abstract_canon_sha256":"99f079ca8ee3082e4eef44385bfcb54c2af926727b299e6459ef1ae688ba36fb"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:54:56.730152Z","signature_b64":"X1GvbKC57WSHcGPMHCxC4Z5o51MK5pqVmw8z/hRa9WvpG2yTT9IbXXLn5a5+ul09s1fbPvsxPOENJR8s5TloAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"a87e747beb46fc447a88c75f919357e891c52dff19cec0db7e8a3e9d5c31145e","last_reissued_at":"2026-07-05T10:54:56.729647Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:54:56.729647Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"GMAR: Gradient-Driven Multi-Head Attention Rollout for Vision Transformer Interpretability","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Gangjae Jang, Haesol Park, Sehyeong Jo","submitted_at":"2025-04-28T01:58:39Z","abstract_excerpt":"The Vision Transformer (ViT) has made significant advancements in computer vision, utilizing self-attention mechanisms to achieve state-of-the-art performance across various tasks, including image classification, object detection, and segmentation. Its architectural flexibility and capabilities have made it a preferred choice among researchers and practitioners. However, the intricate multi-head attention mechanism of ViT presents significant challenges to interpretability, as the underlying prediction process remains opaque. A critical limitation arises from an observation commonly noted in t"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2504.19414","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/2504.19414/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":"2504.19414","created_at":"2026-07-05T10:54:56.729704+00:00"},{"alias_kind":"arxiv_version","alias_value":"2504.19414v1","created_at":"2026-07-05T10:54:56.729704+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2504.19414","created_at":"2026-07-05T10:54:56.729704+00:00"},{"alias_kind":"pith_short_12","alias_value":"VB7HI67LI36E","created_at":"2026-07-05T10:54:56.729704+00:00"},{"alias_kind":"pith_short_16","alias_value":"VB7HI67LI36EI6UI","created_at":"2026-07-05T10:54:56.729704+00:00"},{"alias_kind":"pith_short_8","alias_value":"VB7HI67L","created_at":"2026-07-05T10:54:56.729704+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2601.00014","citing_title":"Modeling Day-Long ECG Signals to Predict Heart Failure Risk with Explainable AI","ref_index":32,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/VB7HI67LI36EI6UIY5PZDE2X5C","json":"https://pith.science/pith/VB7HI67LI36EI6UIY5PZDE2X5C.json","graph_json":"https://pith.science/api/pith-number/VB7HI67LI36EI6UIY5PZDE2X5C/graph.json","events_json":"https://pith.science/api/pith-number/VB7HI67LI36EI6UIY5PZDE2X5C/events.json","paper":"https://pith.science/paper/VB7HI67L"},"agent_actions":{"view_html":"https://pith.science/pith/VB7HI67LI36EI6UIY5PZDE2X5C","download_json":"https://pith.science/pith/VB7HI67LI36EI6UIY5PZDE2X5C.json","view_paper":"https://pith.science/paper/VB7HI67L","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2504.19414&json=true","fetch_graph":"https://pith.science/api/pith-number/VB7HI67LI36EI6UIY5PZDE2X5C/graph.json","fetch_events":"https://pith.science/api/pith-number/VB7HI67LI36EI6UIY5PZDE2X5C/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/VB7HI67LI36EI6UIY5PZDE2X5C/action/timestamp_anchor","attest_storage":"https://pith.science/pith/VB7HI67LI36EI6UIY5PZDE2X5C/action/storage_attestation","attest_author":"https://pith.science/pith/VB7HI67LI36EI6UIY5PZDE2X5C/action/author_attestation","sign_citation":"https://pith.science/pith/VB7HI67LI36EI6UIY5PZDE2X5C/action/citation_signature","submit_replication":"https://pith.science/pith/VB7HI67LI36EI6UIY5PZDE2X5C/action/replication_record"}},"created_at":"2026-07-05T10:54:56.729704+00:00","updated_at":"2026-07-05T10:54:56.729704+00:00"}