{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:ZANUH654ERIVIWZZTD77UTDT7F","short_pith_number":"pith:ZANUH654","schema_version":"1.0","canonical_sha256":"c81b43fbbc2451545b3998fffa4c73f948c906e289693fc6360a1fb6350ae7b9","source":{"kind":"arxiv","id":"2504.04550","version":1},"attestation_state":"computed","paper":{"title":"Advancing Egocentric Video Question Answering with Multimodal Large Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG","cs.RO"],"primary_cat":"cs.CV","authors_text":"Alkesh Patel, Vibhav Chitalia, Yinfei Yang","submitted_at":"2025-04-06T16:58:23Z","abstract_excerpt":"Egocentric Video Question Answering (QA) requires models to handle long-horizon temporal reasoning, first-person perspectives, and specialized challenges like frequent camera movement. This paper systematically evaluates both proprietary and open-source Multimodal Large Language Models (MLLMs) on QaEgo4Dv2 - a refined dataset of egocentric videos derived from QaEgo4D. Four popular MLLMs (GPT-4o, Gemini-1.5-Pro, Video-LLaVa-7B and Qwen2-VL-7B-Instruct) are assessed using zero-shot and fine-tuned approaches for both OpenQA and CloseQA settings. We introduce QaEgo4Dv2 to mitigate annotation noise"},"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":"2504.04550","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2025-04-06T16:58:23Z","cross_cats_sorted":["cs.AI","cs.LG","cs.RO"],"title_canon_sha256":"d1de97c4800e27a769a78017538b8fdf8988d4d8b720dc687e050306c568c9a7","abstract_canon_sha256":"9a6ab2ece62720f04de6cae3afd93937c559dcf5c686c072c3634b076e53a67a"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:45:25.887438Z","signature_b64":"EMBr7R19FtSqfwY1UUyqMdlK8swQU2mTX4L3Mu9MIYQx2fk00LbBREzfdzGYgV1uoKpegDX+Aa9P4ALatDrRBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c81b43fbbc2451545b3998fffa4c73f948c906e289693fc6360a1fb6350ae7b9","last_reissued_at":"2026-07-05T10:45:25.887022Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:45:25.887022Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Advancing Egocentric Video Question Answering with Multimodal Large Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG","cs.RO"],"primary_cat":"cs.CV","authors_text":"Alkesh Patel, Vibhav Chitalia, Yinfei Yang","submitted_at":"2025-04-06T16:58:23Z","abstract_excerpt":"Egocentric Video Question Answering (QA) requires models to handle long-horizon temporal reasoning, first-person perspectives, and specialized challenges like frequent camera movement. This paper systematically evaluates both proprietary and open-source Multimodal Large Language Models (MLLMs) on QaEgo4Dv2 - a refined dataset of egocentric videos derived from QaEgo4D. Four popular MLLMs (GPT-4o, Gemini-1.5-Pro, Video-LLaVa-7B and Qwen2-VL-7B-Instruct) are assessed using zero-shot and fine-tuned approaches for both OpenQA and CloseQA settings. We introduce QaEgo4Dv2 to mitigate annotation noise"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2504.04550","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/2504.04550/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":"2504.04550","created_at":"2026-07-05T10:45:25.887080+00:00"},{"alias_kind":"arxiv_version","alias_value":"2504.04550v1","created_at":"2026-07-05T10:45:25.887080+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2504.04550","created_at":"2026-07-05T10:45:25.887080+00:00"},{"alias_kind":"pith_short_12","alias_value":"ZANUH654ERIV","created_at":"2026-07-05T10:45:25.887080+00:00"},{"alias_kind":"pith_short_16","alias_value":"ZANUH654ERIVIWZZ","created_at":"2026-07-05T10:45:25.887080+00:00"},{"alias_kind":"pith_short_8","alias_value":"ZANUH654","created_at":"2026-07-05T10:45:25.887080+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.25842","citing_title":"Graph it first! Enabling Reasoning on Long-form Egocentric Videos through Scene Graphs","ref_index":31,"is_internal_anchor":false},{"citing_arxiv_id":"2606.13141","citing_title":"Rethinking RAG in Long Videos: What to Retrieve and How to Use It?","ref_index":48,"is_internal_anchor":false},{"citing_arxiv_id":"2606.09142","citing_title":"Decoding Pedestrian Crossing Intention from Egocentric Vision via Vision Language Models","ref_index":52,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/ZANUH654ERIVIWZZTD77UTDT7F","json":"https://pith.science/pith/ZANUH654ERIVIWZZTD77UTDT7F.json","graph_json":"https://pith.science/api/pith-number/ZANUH654ERIVIWZZTD77UTDT7F/graph.json","events_json":"https://pith.science/api/pith-number/ZANUH654ERIVIWZZTD77UTDT7F/events.json","paper":"https://pith.science/paper/ZANUH654"},"agent_actions":{"view_html":"https://pith.science/pith/ZANUH654ERIVIWZZTD77UTDT7F","download_json":"https://pith.science/pith/ZANUH654ERIVIWZZTD77UTDT7F.json","view_paper":"https://pith.science/paper/ZANUH654","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2504.04550&json=true","fetch_graph":"https://pith.science/api/pith-number/ZANUH654ERIVIWZZTD77UTDT7F/graph.json","fetch_events":"https://pith.science/api/pith-number/ZANUH654ERIVIWZZTD77UTDT7F/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ZANUH654ERIVIWZZTD77UTDT7F/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ZANUH654ERIVIWZZTD77UTDT7F/action/storage_attestation","attest_author":"https://pith.science/pith/ZANUH654ERIVIWZZTD77UTDT7F/action/author_attestation","sign_citation":"https://pith.science/pith/ZANUH654ERIVIWZZTD77UTDT7F/action/citation_signature","submit_replication":"https://pith.science/pith/ZANUH654ERIVIWZZTD77UTDT7F/action/replication_record"}},"created_at":"2026-07-05T10:45:25.887080+00:00","updated_at":"2026-07-05T10:45:25.887080+00:00"}