{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:KDTT5IV4SHGRQF4ZUKULRFBO5A","short_pith_number":"pith:KDTT5IV4","schema_version":"1.0","canonical_sha256":"50e73ea2bc91cd181799a2a8b8942ee83380bc99887d332376067cca80554033","source":{"kind":"arxiv","id":"2406.09462","version":1},"attestation_state":"computed","paper":{"title":"SViTT-Ego: A Sparse Video-Text Transformer for Egocentric Video","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Hector A. Valdez, Kyle Min, Subarna Tripathi","submitted_at":"2024-06-13T03:57:38Z","abstract_excerpt":"Pretraining egocentric vision-language models has become essential to improving downstream egocentric video-text tasks. These egocentric foundation models commonly use the transformer architecture. The memory footprint of these models during pretraining can be substantial. Therefore, we pretrain SViTT-Ego, the first sparse egocentric video-text transformer model integrating edge and node sparsification. We pretrain on the EgoClip dataset and incorporate the egocentric-friendly objective EgoNCE, instead of the frequently used InfoNCE. Most notably, SViTT-Ego obtains a +2.8% gain on EgoMCQ (intr"},"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":"2406.09462","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CV","submitted_at":"2024-06-13T03:57:38Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"9d83d38ab4e2c7537934db1f26cf20f80d79f2fd7004f224e927567764f038ba","abstract_canon_sha256":"e5eb56a97008b4dba964896b010c034a2f7ece51f7188d39d1d77061c137d934"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:31:54.405895Z","signature_b64":"oPXK6If37UDQpmneb8OJuukx/1EG4/kUr9ItHeKzFsem0gqadq+6oZ7ifMWUirqgl247j8z0jV2G07TCSdQFBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"50e73ea2bc91cd181799a2a8b8942ee83380bc99887d332376067cca80554033","last_reissued_at":"2026-07-05T08:31:54.405372Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:31:54.405372Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"SViTT-Ego: A Sparse Video-Text Transformer for Egocentric Video","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Hector A. Valdez, Kyle Min, Subarna Tripathi","submitted_at":"2024-06-13T03:57:38Z","abstract_excerpt":"Pretraining egocentric vision-language models has become essential to improving downstream egocentric video-text tasks. These egocentric foundation models commonly use the transformer architecture. The memory footprint of these models during pretraining can be substantial. Therefore, we pretrain SViTT-Ego, the first sparse egocentric video-text transformer model integrating edge and node sparsification. We pretrain on the EgoClip dataset and incorporate the egocentric-friendly objective EgoNCE, instead of the frequently used InfoNCE. Most notably, SViTT-Ego obtains a +2.8% gain on EgoMCQ (intr"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2406.09462","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/2406.09462/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":"2406.09462","created_at":"2026-07-05T08:31:54.405442+00:00"},{"alias_kind":"arxiv_version","alias_value":"2406.09462v1","created_at":"2026-07-05T08:31:54.405442+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2406.09462","created_at":"2026-07-05T08:31:54.405442+00:00"},{"alias_kind":"pith_short_12","alias_value":"KDTT5IV4SHGR","created_at":"2026-07-05T08:31:54.405442+00:00"},{"alias_kind":"pith_short_16","alias_value":"KDTT5IV4SHGRQF4Z","created_at":"2026-07-05T08:31:54.405442+00:00"},{"alias_kind":"pith_short_8","alias_value":"KDTT5IV4","created_at":"2026-07-05T08:31:54.405442+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2506.14356","citing_title":"EVA02-AT: Egocentric Video-Language Understanding with Spatial-Temporal Rotary Positional Embeddings and Symmetric Optimization","ref_index":51,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/KDTT5IV4SHGRQF4ZUKULRFBO5A","json":"https://pith.science/pith/KDTT5IV4SHGRQF4ZUKULRFBO5A.json","graph_json":"https://pith.science/api/pith-number/KDTT5IV4SHGRQF4ZUKULRFBO5A/graph.json","events_json":"https://pith.science/api/pith-number/KDTT5IV4SHGRQF4ZUKULRFBO5A/events.json","paper":"https://pith.science/paper/KDTT5IV4"},"agent_actions":{"view_html":"https://pith.science/pith/KDTT5IV4SHGRQF4ZUKULRFBO5A","download_json":"https://pith.science/pith/KDTT5IV4SHGRQF4ZUKULRFBO5A.json","view_paper":"https://pith.science/paper/KDTT5IV4","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2406.09462&json=true","fetch_graph":"https://pith.science/api/pith-number/KDTT5IV4SHGRQF4ZUKULRFBO5A/graph.json","fetch_events":"https://pith.science/api/pith-number/KDTT5IV4SHGRQF4ZUKULRFBO5A/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/KDTT5IV4SHGRQF4ZUKULRFBO5A/action/timestamp_anchor","attest_storage":"https://pith.science/pith/KDTT5IV4SHGRQF4ZUKULRFBO5A/action/storage_attestation","attest_author":"https://pith.science/pith/KDTT5IV4SHGRQF4ZUKULRFBO5A/action/author_attestation","sign_citation":"https://pith.science/pith/KDTT5IV4SHGRQF4ZUKULRFBO5A/action/citation_signature","submit_replication":"https://pith.science/pith/KDTT5IV4SHGRQF4ZUKULRFBO5A/action/replication_record"}},"created_at":"2026-07-05T08:31:54.405442+00:00","updated_at":"2026-07-05T08:31:54.405442+00:00"}