{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:SBRAKSJXMKHERCMNVCNPLYWYBZ","short_pith_number":"pith:SBRAKSJX","schema_version":"1.0","canonical_sha256":"9062054937628e48898da89af5e2d80e67ceab2ad86ccf7f9d1eb7868c7fa615","source":{"kind":"arxiv","id":"2506.14356","version":1},"attestation_state":"computed","paper":{"title":"EVA02-AT: Egocentric Video-Language Understanding with Spatial-Temporal Rotary Positional Embeddings and Symmetric Optimization","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Lap-Pui Chau, Xiaoqi Wang, Yi Wang","submitted_at":"2025-06-17T09:51:51Z","abstract_excerpt":"Egocentric video-language understanding demands both high efficiency and accurate spatial-temporal modeling. Existing approaches face three key challenges: 1) Excessive pre-training cost arising from multi-stage pre-training pipelines, 2) Ineffective spatial-temporal encoding due to manually split 3D rotary positional embeddings that hinder feature interactions, and 3) Imprecise learning objectives in soft-label multi-instance retrieval, which neglect negative pair correlations. In this paper, we introduce EVA02-AT, a suite of EVA02-based video-language foundation models tailored to egocentric"},"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":"2506.14356","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2025-06-17T09:51:51Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"423e234f030a449e4e7238da5e0dd989e94696172519bc9579d1f877bb9fa2f5","abstract_canon_sha256":"fc87a4f96f9eb382c4bb4f83d057a505ee24a6bd19466cb946ad165aa109d688"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:22:53.964024Z","signature_b64":"L+sm/Mw238zY6Y5WhrIS6aB926UbPrkHQWiI0q1wXkZsxRljRJ0VWjoMoxjBgqBQXyfVRWsB5GxE8PIVhup2Cg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9062054937628e48898da89af5e2d80e67ceab2ad86ccf7f9d1eb7868c7fa615","last_reissued_at":"2026-07-05T11:22:53.963643Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:22:53.963643Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"EVA02-AT: Egocentric Video-Language Understanding with Spatial-Temporal Rotary Positional Embeddings and Symmetric Optimization","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Lap-Pui Chau, Xiaoqi Wang, Yi Wang","submitted_at":"2025-06-17T09:51:51Z","abstract_excerpt":"Egocentric video-language understanding demands both high efficiency and accurate spatial-temporal modeling. Existing approaches face three key challenges: 1) Excessive pre-training cost arising from multi-stage pre-training pipelines, 2) Ineffective spatial-temporal encoding due to manually split 3D rotary positional embeddings that hinder feature interactions, and 3) Imprecise learning objectives in soft-label multi-instance retrieval, which neglect negative pair correlations. In this paper, we introduce EVA02-AT, a suite of EVA02-based video-language foundation models tailored to egocentric"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.14356","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/2506.14356/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":"2506.14356","created_at":"2026-07-05T11:22:53.963691+00:00"},{"alias_kind":"arxiv_version","alias_value":"2506.14356v1","created_at":"2026-07-05T11:22:53.963691+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.14356","created_at":"2026-07-05T11:22:53.963691+00:00"},{"alias_kind":"pith_short_12","alias_value":"SBRAKSJXMKHE","created_at":"2026-07-05T11:22:53.963691+00:00"},{"alias_kind":"pith_short_16","alias_value":"SBRAKSJXMKHERCMN","created_at":"2026-07-05T11:22:53.963691+00:00"},{"alias_kind":"pith_short_8","alias_value":"SBRAKSJX","created_at":"2026-07-05T11:22:53.963691+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.14742","citing_title":"EARL: Towards a Unified Analysis-Guided Reinforcement Learning Framework for Egocentric Interaction Reasoning and Pixel Grounding","ref_index":16,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/SBRAKSJXMKHERCMNVCNPLYWYBZ","json":"https://pith.science/pith/SBRAKSJXMKHERCMNVCNPLYWYBZ.json","graph_json":"https://pith.science/api/pith-number/SBRAKSJXMKHERCMNVCNPLYWYBZ/graph.json","events_json":"https://pith.science/api/pith-number/SBRAKSJXMKHERCMNVCNPLYWYBZ/events.json","paper":"https://pith.science/paper/SBRAKSJX"},"agent_actions":{"view_html":"https://pith.science/pith/SBRAKSJXMKHERCMNVCNPLYWYBZ","download_json":"https://pith.science/pith/SBRAKSJXMKHERCMNVCNPLYWYBZ.json","view_paper":"https://pith.science/paper/SBRAKSJX","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2506.14356&json=true","fetch_graph":"https://pith.science/api/pith-number/SBRAKSJXMKHERCMNVCNPLYWYBZ/graph.json","fetch_events":"https://pith.science/api/pith-number/SBRAKSJXMKHERCMNVCNPLYWYBZ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/SBRAKSJXMKHERCMNVCNPLYWYBZ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/SBRAKSJXMKHERCMNVCNPLYWYBZ/action/storage_attestation","attest_author":"https://pith.science/pith/SBRAKSJXMKHERCMNVCNPLYWYBZ/action/author_attestation","sign_citation":"https://pith.science/pith/SBRAKSJXMKHERCMNVCNPLYWYBZ/action/citation_signature","submit_replication":"https://pith.science/pith/SBRAKSJXMKHERCMNVCNPLYWYBZ/action/replication_record"}},"created_at":"2026-07-05T11:22:53.963691+00:00","updated_at":"2026-07-05T11:22:53.963691+00:00"}