{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:SQ4WT45JT2ZU3GSTYRMIAN4WOL","short_pith_number":"pith:SQ4WT45J","schema_version":"1.0","canonical_sha256":"943969f3a99eb34d9a53c45880379672ea782f2698eadf87be61346b423ba905","source":{"kind":"arxiv","id":"2204.01680","version":2},"attestation_state":"computed","paper":{"title":"TALLFormer: Temporal Action Localization with a Long-memory Transformer","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Feng Cheng, Gedas Bertasius","submitted_at":"2022-04-04T17:51:20Z","abstract_excerpt":"Most modern approaches in temporal action localization divide this problem into two parts: (i) short-term feature extraction and (ii) long-range temporal boundary localization. Due to the high GPU memory cost caused by processing long untrimmed videos, many methods sacrifice the representational power of the short-term feature extractor by either freezing the backbone or using a small spatial video resolution. This issue becomes even worse with the recent video transformer models, many of which have quadratic memory complexity. To address these issues, we propose TALLFormer, a memory-efficient"},"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":"2204.01680","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2022-04-04T17:51:20Z","cross_cats_sorted":[],"title_canon_sha256":"a58c158f34362a216d75d0b93d20be80faa0ed2b019a9debe7d1bc8b705fe038","abstract_canon_sha256":"fb63ee8de3a647c5a7d542a6fedae79fe7dc4fd7a964daeabe4dd209d86e2ce4"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:44:00.858206Z","signature_b64":"7g/6fP6Dyif3Jxp6CME3B40JIz+/zm5Tp2QlAAjx7L+pIm/3blJGj5gXzurmZY7zTO21rJPwMFeauBZ2DIyQCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"943969f3a99eb34d9a53c45880379672ea782f2698eadf87be61346b423ba905","last_reissued_at":"2026-07-05T04:44:00.857785Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:44:00.857785Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"TALLFormer: Temporal Action Localization with a Long-memory Transformer","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Feng Cheng, Gedas Bertasius","submitted_at":"2022-04-04T17:51:20Z","abstract_excerpt":"Most modern approaches in temporal action localization divide this problem into two parts: (i) short-term feature extraction and (ii) long-range temporal boundary localization. Due to the high GPU memory cost caused by processing long untrimmed videos, many methods sacrifice the representational power of the short-term feature extractor by either freezing the backbone or using a small spatial video resolution. This issue becomes even worse with the recent video transformer models, many of which have quadratic memory complexity. To address these issues, we propose TALLFormer, a memory-efficient"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2204.01680","kind":"arxiv","version":2},"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/2204.01680/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":"2204.01680","created_at":"2026-07-05T04:44:00.857843+00:00"},{"alias_kind":"arxiv_version","alias_value":"2204.01680v2","created_at":"2026-07-05T04:44:00.857843+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2204.01680","created_at":"2026-07-05T04:44:00.857843+00:00"},{"alias_kind":"pith_short_12","alias_value":"SQ4WT45JT2ZU","created_at":"2026-07-05T04:44:00.857843+00:00"},{"alias_kind":"pith_short_16","alias_value":"SQ4WT45JT2ZU3GST","created_at":"2026-07-05T04:44:00.857843+00:00"},{"alias_kind":"pith_short_8","alias_value":"SQ4WT45J","created_at":"2026-07-05T04:44:00.857843+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2411.14688","citing_title":"Whats in a Video: Factorized Autoregressive Decoding for Online Dense Video Captioning","ref_index":11,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/SQ4WT45JT2ZU3GSTYRMIAN4WOL","json":"https://pith.science/pith/SQ4WT45JT2ZU3GSTYRMIAN4WOL.json","graph_json":"https://pith.science/api/pith-number/SQ4WT45JT2ZU3GSTYRMIAN4WOL/graph.json","events_json":"https://pith.science/api/pith-number/SQ4WT45JT2ZU3GSTYRMIAN4WOL/events.json","paper":"https://pith.science/paper/SQ4WT45J"},"agent_actions":{"view_html":"https://pith.science/pith/SQ4WT45JT2ZU3GSTYRMIAN4WOL","download_json":"https://pith.science/pith/SQ4WT45JT2ZU3GSTYRMIAN4WOL.json","view_paper":"https://pith.science/paper/SQ4WT45J","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2204.01680&json=true","fetch_graph":"https://pith.science/api/pith-number/SQ4WT45JT2ZU3GSTYRMIAN4WOL/graph.json","fetch_events":"https://pith.science/api/pith-number/SQ4WT45JT2ZU3GSTYRMIAN4WOL/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/SQ4WT45JT2ZU3GSTYRMIAN4WOL/action/timestamp_anchor","attest_storage":"https://pith.science/pith/SQ4WT45JT2ZU3GSTYRMIAN4WOL/action/storage_attestation","attest_author":"https://pith.science/pith/SQ4WT45JT2ZU3GSTYRMIAN4WOL/action/author_attestation","sign_citation":"https://pith.science/pith/SQ4WT45JT2ZU3GSTYRMIAN4WOL/action/citation_signature","submit_replication":"https://pith.science/pith/SQ4WT45JT2ZU3GSTYRMIAN4WOL/action/replication_record"}},"created_at":"2026-07-05T04:44:00.857843+00:00","updated_at":"2026-07-05T04:44:00.857843+00:00"}