{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:EOEDF7EXBDLUE2PFLYQJCZMFMU","short_pith_number":"pith:EOEDF7EX","schema_version":"1.0","canonical_sha256":"238832fc9708d74269e55e209165856507d110dcdcd49435896adfced0a2a240","source":{"kind":"arxiv","id":"2605.19075","version":1},"attestation_state":"computed","paper":{"title":"CRAFT: Critic-Refined Adaptive Key-Frame Targeting for Multimodal Video Question Answering","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Abdul Wasi, Akhil Gorugantu, David Doermann, Mahesh Bhosale, Pengyu Yan, Vishvesh Trivedi","submitted_at":"2026-05-18T20:01:05Z","abstract_excerpt":"Grounded multi-video question answering over real-world news events requires systems to surface query-relevant evidence across heterogeneous video archives while attributing every claim to its supporting source. We introduce CRAFT (Critic-Refined Adaptive Key-Frame Targeting), a query-conditioned pipeline that combines dynamic keyframe selection, per-video ASR with multilingual fallback, and a hybrid critic loop to iteratively verify and repair claims before consolidation. The pipeline integrates UNLI temporal entailment, DeBERTa-v3 cross-claim screening, and a Llama-3.2-3B adjudicator, with a"},"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":"2605.19075","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-05-18T20:01:05Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"1e820cd0e7f4309efdb4b2cc3567aeee7de51f26a0b74b3ab8416e294a414d32","abstract_canon_sha256":"aa1fbffb7ce2bfcfe19064528f2cec299147f3f0c39828ebe93a0aafe1523291"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-20T01:05:26.175206Z","signature_b64":"m2GFmYY9ZhHW0+d427W1dLkUQAjhJyJjOAV7kwKqHDysARN3Q3jevRq8OQwlnyU0UEyc2SFCA9wlhJ1V21XgAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"238832fc9708d74269e55e209165856507d110dcdcd49435896adfced0a2a240","last_reissued_at":"2026-05-20T01:05:26.174610Z","signature_status":"signed_v1","first_computed_at":"2026-05-20T01:05:26.174610Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"CRAFT: Critic-Refined Adaptive Key-Frame Targeting for Multimodal Video Question Answering","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Abdul Wasi, Akhil Gorugantu, David Doermann, Mahesh Bhosale, Pengyu Yan, Vishvesh Trivedi","submitted_at":"2026-05-18T20:01:05Z","abstract_excerpt":"Grounded multi-video question answering over real-world news events requires systems to surface query-relevant evidence across heterogeneous video archives while attributing every claim to its supporting source. We introduce CRAFT (Critic-Refined Adaptive Key-Frame Targeting), a query-conditioned pipeline that combines dynamic keyframe selection, per-video ASR with multilingual fallback, and a hybrid critic loop to iteratively verify and repair claims before consolidation. The pipeline integrates UNLI temporal entailment, DeBERTa-v3 cross-claim screening, and a Llama-3.2-3B adjudicator, with a"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.19075","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/2605.19075/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":"2605.19075","created_at":"2026-05-20T01:05:26.174701+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.19075v1","created_at":"2026-05-20T01:05:26.174701+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.19075","created_at":"2026-05-20T01:05:26.174701+00:00"},{"alias_kind":"pith_short_12","alias_value":"EOEDF7EXBDLU","created_at":"2026-05-20T01:05:26.174701+00:00"},{"alias_kind":"pith_short_16","alias_value":"EOEDF7EXBDLUE2PF","created_at":"2026-05-20T01:05:26.174701+00:00"},{"alias_kind":"pith_short_8","alias_value":"EOEDF7EX","created_at":"2026-05-20T01:05:26.174701+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/EOEDF7EXBDLUE2PFLYQJCZMFMU","json":"https://pith.science/pith/EOEDF7EXBDLUE2PFLYQJCZMFMU.json","graph_json":"https://pith.science/api/pith-number/EOEDF7EXBDLUE2PFLYQJCZMFMU/graph.json","events_json":"https://pith.science/api/pith-number/EOEDF7EXBDLUE2PFLYQJCZMFMU/events.json","paper":"https://pith.science/paper/EOEDF7EX"},"agent_actions":{"view_html":"https://pith.science/pith/EOEDF7EXBDLUE2PFLYQJCZMFMU","download_json":"https://pith.science/pith/EOEDF7EXBDLUE2PFLYQJCZMFMU.json","view_paper":"https://pith.science/paper/EOEDF7EX","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.19075&json=true","fetch_graph":"https://pith.science/api/pith-number/EOEDF7EXBDLUE2PFLYQJCZMFMU/graph.json","fetch_events":"https://pith.science/api/pith-number/EOEDF7EXBDLUE2PFLYQJCZMFMU/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/EOEDF7EXBDLUE2PFLYQJCZMFMU/action/timestamp_anchor","attest_storage":"https://pith.science/pith/EOEDF7EXBDLUE2PFLYQJCZMFMU/action/storage_attestation","attest_author":"https://pith.science/pith/EOEDF7EXBDLUE2PFLYQJCZMFMU/action/author_attestation","sign_citation":"https://pith.science/pith/EOEDF7EXBDLUE2PFLYQJCZMFMU/action/citation_signature","submit_replication":"https://pith.science/pith/EOEDF7EXBDLUE2PFLYQJCZMFMU/action/replication_record"}},"created_at":"2026-05-20T01:05:26.174701+00:00","updated_at":"2026-05-20T01:05:26.174701+00:00"}