{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:LGE7PQ342OR7AW7XYY7PZUA7DV","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"43adb24f2f1c4c80e11ff73ee54cfbcfd4b191735f40fe3b25f6f4f3b6895e58","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2023-01-05T08:21:01Z","title_canon_sha256":"5559cdc2e7bf9438bb10efb47f2c07d7018a7aaababed8508638d1671bb576b6"},"schema_version":"1.0","source":{"id":"2301.01953","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2301.01953","created_at":"2026-07-05T05:49:32Z"},{"alias_kind":"arxiv_version","alias_value":"2301.01953v3","created_at":"2026-07-05T05:49:32Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2301.01953","created_at":"2026-07-05T05:49:32Z"},{"alias_kind":"pith_short_12","alias_value":"LGE7PQ342OR7","created_at":"2026-07-05T05:49:32Z"},{"alias_kind":"pith_short_16","alias_value":"LGE7PQ342OR7AW7X","created_at":"2026-07-05T05:49:32Z"},{"alias_kind":"pith_short_8","alias_value":"LGE7PQ34","created_at":"2026-07-05T05:49:32Z"}],"graph_snapshots":[{"event_id":"sha256:28bd57b643d800b5f66163c20cd38fd122c7da0c7caf474099c3e84c79dace52","target":"graph","created_at":"2026-07-05T05:49:32Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2301.01953/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"In a video, an object usually appears as the trajectory, i.e., it spans over a few spatial but longer temporal patches, that contains abundant spatiotemporal contexts. However, modern Video-Language BERTs (VDL-BERTs) neglect this trajectory characteristic that they usually follow image-language BERTs (IL-BERTs) to deploy the patch-to-word (P2W) attention that may over-exploit trivial spatial contexts and neglect significant temporal contexts. To amend this, we propose a novel TW-BERT to learn Trajectory-Word alignment by a newly designed trajectory-to-word (T2W) attention for solving video-lan","authors_text":"Chenliang Li, Fei Huang, Haiyang Xu, Hanwang Zhang, Ming Yan, Qinghao Ye, Songfang Huang, Xu Yang, Yu Zhang, Zhangzikang Li","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2023-01-05T08:21:01Z","title":"Learning Trajectory-Word Alignments for Video-Language Tasks"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2301.01953","kind":"arxiv","version":3},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:9fdc871c011480713cd22863601ef17ed6bb237692496a92b201e879718d4f23","target":"record","created_at":"2026-07-05T05:49:32Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"43adb24f2f1c4c80e11ff73ee54cfbcfd4b191735f40fe3b25f6f4f3b6895e58","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2023-01-05T08:21:01Z","title_canon_sha256":"5559cdc2e7bf9438bb10efb47f2c07d7018a7aaababed8508638d1671bb576b6"},"schema_version":"1.0","source":{"id":"2301.01953","kind":"arxiv","version":3}},"canonical_sha256":"5989f7c37cd3a3f05bf7c63efcd01f1d4a2ab3ba1da8cfb53f906749b922da33","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"5989f7c37cd3a3f05bf7c63efcd01f1d4a2ab3ba1da8cfb53f906749b922da33","first_computed_at":"2026-07-05T05:49:32.253101Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T05:49:32.253101Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"IKjeUlVxD2qHOdMy624x2Ct3TZ+TI2fEsXWklh0Ep9XU8LTJss58qkfil6uQrS2Axlopd9o/f4NhKDesGdYcAA==","signature_status":"signed_v1","signed_at":"2026-07-05T05:49:32.253478Z","signed_message":"canonical_sha256_bytes"},"source_id":"2301.01953","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:9fdc871c011480713cd22863601ef17ed6bb237692496a92b201e879718d4f23","sha256:28bd57b643d800b5f66163c20cd38fd122c7da0c7caf474099c3e84c79dace52"],"state_sha256":"783551fff43cda0540fe541bd1dba7fc36b339842d73c61e6b7ae3215928ca2a"}