{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:NQAJRPXFRUCHP4LQ3KXLVEPNSL","short_pith_number":"pith:NQAJRPXF","schema_version":"1.0","canonical_sha256":"6c0098bee58d0477f170daaeba91ed92cf777bffbcfd0ca492a49056c6154acc","source":{"kind":"arxiv","id":"2503.19120","version":1},"attestation_state":"computed","paper":{"title":"Where is this coming from? Making groundedness count in the evaluation of Document VQA models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Armineh Nourbakhsh, Carolyn Rose, Pranav Shetty, Sameena Shah, Siddharth Parekh, Zhao Jin","submitted_at":"2025-03-24T20:14:46Z","abstract_excerpt":"Document Visual Question Answering (VQA) models have evolved at an impressive rate over the past few years, coming close to or matching human performance on some benchmarks. We argue that common evaluation metrics used by popular benchmarks do not account for the semantic and multimodal groundedness of a model's outputs. As a result, hallucinations and major semantic errors are treated the same way as well-grounded outputs, and the evaluation scores do not reflect the reasoning capabilities of the model. In response, we propose a new evaluation methodology that accounts for the groundedness of"},"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":"2503.19120","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-03-24T20:14:46Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"6299b2e6b7130338825ea259f56bbb18a340da9ab9736fcf6a020a958a1bdc99","abstract_canon_sha256":"6850edea178f4e38289fe01d4da28ea448a1af7fe2b14eb9f55ac336eff1bb7a"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:38:49.925266Z","signature_b64":"zB/9CIfQCdclGJcwT4N+x5H+4VjwEUfHkNHb4pgb+kiI1tTq2bqR2/QuDgYNlU8IXmzJtzWt5pb2zfobD6jEDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"6c0098bee58d0477f170daaeba91ed92cf777bffbcfd0ca492a49056c6154acc","last_reissued_at":"2026-07-05T10:38:49.924767Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:38:49.924767Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Where is this coming from? Making groundedness count in the evaluation of Document VQA models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Armineh Nourbakhsh, Carolyn Rose, Pranav Shetty, Sameena Shah, Siddharth Parekh, Zhao Jin","submitted_at":"2025-03-24T20:14:46Z","abstract_excerpt":"Document Visual Question Answering (VQA) models have evolved at an impressive rate over the past few years, coming close to or matching human performance on some benchmarks. We argue that common evaluation metrics used by popular benchmarks do not account for the semantic and multimodal groundedness of a model's outputs. As a result, hallucinations and major semantic errors are treated the same way as well-grounded outputs, and the evaluation scores do not reflect the reasoning capabilities of the model. In response, we propose a new evaluation methodology that accounts for the groundedness of"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2503.19120","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/2503.19120/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":"2503.19120","created_at":"2026-07-05T10:38:49.924821+00:00"},{"alias_kind":"arxiv_version","alias_value":"2503.19120v1","created_at":"2026-07-05T10:38:49.924821+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2503.19120","created_at":"2026-07-05T10:38:49.924821+00:00"},{"alias_kind":"pith_short_12","alias_value":"NQAJRPXFRUCH","created_at":"2026-07-05T10:38:49.924821+00:00"},{"alias_kind":"pith_short_16","alias_value":"NQAJRPXFRUCHP4LQ","created_at":"2026-07-05T10:38:49.924821+00:00"},{"alias_kind":"pith_short_8","alias_value":"NQAJRPXF","created_at":"2026-07-05T10:38:49.924821+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/NQAJRPXFRUCHP4LQ3KXLVEPNSL","json":"https://pith.science/pith/NQAJRPXFRUCHP4LQ3KXLVEPNSL.json","graph_json":"https://pith.science/api/pith-number/NQAJRPXFRUCHP4LQ3KXLVEPNSL/graph.json","events_json":"https://pith.science/api/pith-number/NQAJRPXFRUCHP4LQ3KXLVEPNSL/events.json","paper":"https://pith.science/paper/NQAJRPXF"},"agent_actions":{"view_html":"https://pith.science/pith/NQAJRPXFRUCHP4LQ3KXLVEPNSL","download_json":"https://pith.science/pith/NQAJRPXFRUCHP4LQ3KXLVEPNSL.json","view_paper":"https://pith.science/paper/NQAJRPXF","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2503.19120&json=true","fetch_graph":"https://pith.science/api/pith-number/NQAJRPXFRUCHP4LQ3KXLVEPNSL/graph.json","fetch_events":"https://pith.science/api/pith-number/NQAJRPXFRUCHP4LQ3KXLVEPNSL/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/NQAJRPXFRUCHP4LQ3KXLVEPNSL/action/timestamp_anchor","attest_storage":"https://pith.science/pith/NQAJRPXFRUCHP4LQ3KXLVEPNSL/action/storage_attestation","attest_author":"https://pith.science/pith/NQAJRPXFRUCHP4LQ3KXLVEPNSL/action/author_attestation","sign_citation":"https://pith.science/pith/NQAJRPXFRUCHP4LQ3KXLVEPNSL/action/citation_signature","submit_replication":"https://pith.science/pith/NQAJRPXFRUCHP4LQ3KXLVEPNSL/action/replication_record"}},"created_at":"2026-07-05T10:38:49.924821+00:00","updated_at":"2026-07-05T10:38:49.924821+00:00"}