{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:3T44ORLSGJXYK6QY3BKGJYQ4OM","short_pith_number":"pith:3T44ORLS","schema_version":"1.0","canonical_sha256":"dcf9c74572326f857a18d85464e21c73368c7f5ae98224bad157367ecbbe0c3c","source":{"kind":"arxiv","id":"2105.02626","version":1},"attestation_state":"computed","paper":{"title":"A First Look: Towards Explainable TextVQA Models via Visual and Textual Explanations","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CL","cs.LG"],"primary_cat":"cs.CV","authors_text":"Karen Hovsepian, Mingwei Shen, Varun Nagaraj Rao, Xingjian Zhen","submitted_at":"2021-04-29T00:36:17Z","abstract_excerpt":"Explainable deep learning models are advantageous in many situations. Prior work mostly provide unimodal explanations through post-hoc approaches not part of the original system design. Explanation mechanisms also ignore useful textual information present in images. In this paper, we propose MTXNet, an end-to-end trainable multimodal architecture to generate multimodal explanations, which focuses on the text in the image. We curate a novel dataset TextVQA-X, containing ground truth visual and multi-reference textual explanations that can be leveraged during both training and evaluation. We 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":"2105.02626","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2021-04-29T00:36:17Z","cross_cats_sorted":["cs.AI","cs.CL","cs.LG"],"title_canon_sha256":"1b3f32d8192c1c8e8e3a099b95e6ccd01a82e36265cd0ddc85f902074eaf227f","abstract_canon_sha256":"eb7e08578e247c6074778f31b62cc6fb650c185fc4ee9c6d3e1b51e60e6b2a20"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T02:38:03.175161Z","signature_b64":"55/sF3QxK07myXUEInNeVBn8UnUkhrNC1GDwtPqFqLsR2woeTiSevUF4ZBF2Or+Y6yzCpJCy+Sqr2IElk1CzAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"dcf9c74572326f857a18d85464e21c73368c7f5ae98224bad157367ecbbe0c3c","last_reissued_at":"2026-07-05T02:38:03.174767Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T02:38:03.174767Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"A First Look: Towards Explainable TextVQA Models via Visual and Textual Explanations","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CL","cs.LG"],"primary_cat":"cs.CV","authors_text":"Karen Hovsepian, Mingwei Shen, Varun Nagaraj Rao, Xingjian Zhen","submitted_at":"2021-04-29T00:36:17Z","abstract_excerpt":"Explainable deep learning models are advantageous in many situations. Prior work mostly provide unimodal explanations through post-hoc approaches not part of the original system design. Explanation mechanisms also ignore useful textual information present in images. In this paper, we propose MTXNet, an end-to-end trainable multimodal architecture to generate multimodal explanations, which focuses on the text in the image. We curate a novel dataset TextVQA-X, containing ground truth visual and multi-reference textual explanations that can be leveraged during both training and evaluation. We the"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2105.02626","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/2105.02626/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":"2105.02626","created_at":"2026-07-05T02:38:03.174817+00:00"},{"alias_kind":"arxiv_version","alias_value":"2105.02626v1","created_at":"2026-07-05T02:38:03.174817+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2105.02626","created_at":"2026-07-05T02:38:03.174817+00:00"},{"alias_kind":"pith_short_12","alias_value":"3T44ORLSGJXY","created_at":"2026-07-05T02:38:03.174817+00:00"},{"alias_kind":"pith_short_16","alias_value":"3T44ORLSGJXYK6QY","created_at":"2026-07-05T02:38:03.174817+00:00"},{"alias_kind":"pith_short_8","alias_value":"3T44ORLS","created_at":"2026-07-05T02:38:03.174817+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2411.19187","citing_title":"Beyond Logit Lens: Contextual Embeddings for Robust Hallucination Detection & Grounding in VLMs","ref_index":31,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/3T44ORLSGJXYK6QY3BKGJYQ4OM","json":"https://pith.science/pith/3T44ORLSGJXYK6QY3BKGJYQ4OM.json","graph_json":"https://pith.science/api/pith-number/3T44ORLSGJXYK6QY3BKGJYQ4OM/graph.json","events_json":"https://pith.science/api/pith-number/3T44ORLSGJXYK6QY3BKGJYQ4OM/events.json","paper":"https://pith.science/paper/3T44ORLS"},"agent_actions":{"view_html":"https://pith.science/pith/3T44ORLSGJXYK6QY3BKGJYQ4OM","download_json":"https://pith.science/pith/3T44ORLSGJXYK6QY3BKGJYQ4OM.json","view_paper":"https://pith.science/paper/3T44ORLS","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2105.02626&json=true","fetch_graph":"https://pith.science/api/pith-number/3T44ORLSGJXYK6QY3BKGJYQ4OM/graph.json","fetch_events":"https://pith.science/api/pith-number/3T44ORLSGJXYK6QY3BKGJYQ4OM/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/3T44ORLSGJXYK6QY3BKGJYQ4OM/action/timestamp_anchor","attest_storage":"https://pith.science/pith/3T44ORLSGJXYK6QY3BKGJYQ4OM/action/storage_attestation","attest_author":"https://pith.science/pith/3T44ORLSGJXYK6QY3BKGJYQ4OM/action/author_attestation","sign_citation":"https://pith.science/pith/3T44ORLSGJXYK6QY3BKGJYQ4OM/action/citation_signature","submit_replication":"https://pith.science/pith/3T44ORLSGJXYK6QY3BKGJYQ4OM/action/replication_record"}},"created_at":"2026-07-05T02:38:03.174817+00:00","updated_at":"2026-07-05T02:38:03.174817+00:00"}