{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:POHEBDH5FJNAEVWOWO4UGGTQWG","short_pith_number":"pith:POHEBDH5","schema_version":"1.0","canonical_sha256":"7b8e408cfd2a5a0256ceb3b9431a70b1bd759f02ac29006041f191c2c49fe70f","source":{"kind":"arxiv","id":"2409.01156","version":2},"attestation_state":"computed","paper":{"title":"TempMe: Video Temporal Token Merging for Efficient Text-Video Retrieval","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Guiguang Ding, Leqi Shen, Pengzhang Liu, Sicheng Zhao, Tao He, Tianxiang Hao, Yifeng Zhang, Yongjun Bao","submitted_at":"2024-09-02T10:42:30Z","abstract_excerpt":"Most text-video retrieval methods utilize the text-image pre-trained models like CLIP as a backbone. These methods process each sampled frame independently by the image encoder, resulting in high computational overhead and limiting practical deployment. Addressing this, we focus on efficient text-video retrieval by tackling two key challenges: 1. From the perspective of trainable parameters, current parameter-efficient fine-tuning methods incur high inference costs; 2. From the perspective of model complexity, current token compression methods are mainly designed for images to reduce spatial r"},"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":"2409.01156","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2024-09-02T10:42:30Z","cross_cats_sorted":[],"title_canon_sha256":"9859366e6461948756b18a8e2609ed08d203f228bb564246273e2d769a275ddb","abstract_canon_sha256":"159700c832e5242278e44ca344167e0ab138e43b2a5661fdcbcf3206f444fcfe"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:29:22.530136Z","signature_b64":"NaYcoyccQvnfFIK7VKYPDEQfKCi7HkBRtLPx/fGF/MAPJ9pbMSNrfgXNNRX0EaWcYv2QRZ9m4GRW9/3+rbCKCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7b8e408cfd2a5a0256ceb3b9431a70b1bd759f02ac29006041f191c2c49fe70f","last_reissued_at":"2026-07-05T10:29:22.529377Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:29:22.529377Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"TempMe: Video Temporal Token Merging for Efficient Text-Video Retrieval","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Guiguang Ding, Leqi Shen, Pengzhang Liu, Sicheng Zhao, Tao He, Tianxiang Hao, Yifeng Zhang, Yongjun Bao","submitted_at":"2024-09-02T10:42:30Z","abstract_excerpt":"Most text-video retrieval methods utilize the text-image pre-trained models like CLIP as a backbone. These methods process each sampled frame independently by the image encoder, resulting in high computational overhead and limiting practical deployment. Addressing this, we focus on efficient text-video retrieval by tackling two key challenges: 1. From the perspective of trainable parameters, current parameter-efficient fine-tuning methods incur high inference costs; 2. From the perspective of model complexity, current token compression methods are mainly designed for images to reduce spatial r"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2409.01156","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/2409.01156/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":"2409.01156","created_at":"2026-07-05T10:29:22.529477+00:00"},{"alias_kind":"arxiv_version","alias_value":"2409.01156v2","created_at":"2026-07-05T10:29:22.529477+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2409.01156","created_at":"2026-07-05T10:29:22.529477+00:00"},{"alias_kind":"pith_short_12","alias_value":"POHEBDH5FJNA","created_at":"2026-07-05T10:29:22.529477+00:00"},{"alias_kind":"pith_short_16","alias_value":"POHEBDH5FJNAEVWO","created_at":"2026-07-05T10:29:22.529477+00:00"},{"alias_kind":"pith_short_8","alias_value":"POHEBDH5","created_at":"2026-07-05T10:29:22.529477+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":5,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.29350","citing_title":"Fast Enough to Act: Spatio-Temporal Visual Token Merging for Low-Latency Robotic VLMs and VLAs","ref_index":41,"is_internal_anchor":false},{"citing_arxiv_id":"2509.02560","citing_title":"FastVGGT: Training-Free Acceleration of Visual Geometry Transformer","ref_index":20,"is_internal_anchor":false},{"citing_arxiv_id":"2602.20913","citing_title":"LongVideo-R1: Smart Navigation for Low-cost Long Video Understanding","ref_index":38,"is_internal_anchor":false},{"citing_arxiv_id":"2603.01400","citing_title":"Token Reduction via Local and Global Contexts Optimization for Efficient Video Large Language Models","ref_index":46,"is_internal_anchor":false},{"citing_arxiv_id":"2605.00826","citing_title":"Understanding the Performance Plateau in Text-to-Video Retrieval: A Comprehensive Empirical and Linguistic Analysis","ref_index":67,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/POHEBDH5FJNAEVWOWO4UGGTQWG","json":"https://pith.science/pith/POHEBDH5FJNAEVWOWO4UGGTQWG.json","graph_json":"https://pith.science/api/pith-number/POHEBDH5FJNAEVWOWO4UGGTQWG/graph.json","events_json":"https://pith.science/api/pith-number/POHEBDH5FJNAEVWOWO4UGGTQWG/events.json","paper":"https://pith.science/paper/POHEBDH5"},"agent_actions":{"view_html":"https://pith.science/pith/POHEBDH5FJNAEVWOWO4UGGTQWG","download_json":"https://pith.science/pith/POHEBDH5FJNAEVWOWO4UGGTQWG.json","view_paper":"https://pith.science/paper/POHEBDH5","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2409.01156&json=true","fetch_graph":"https://pith.science/api/pith-number/POHEBDH5FJNAEVWOWO4UGGTQWG/graph.json","fetch_events":"https://pith.science/api/pith-number/POHEBDH5FJNAEVWOWO4UGGTQWG/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/POHEBDH5FJNAEVWOWO4UGGTQWG/action/timestamp_anchor","attest_storage":"https://pith.science/pith/POHEBDH5FJNAEVWOWO4UGGTQWG/action/storage_attestation","attest_author":"https://pith.science/pith/POHEBDH5FJNAEVWOWO4UGGTQWG/action/author_attestation","sign_citation":"https://pith.science/pith/POHEBDH5FJNAEVWOWO4UGGTQWG/action/citation_signature","submit_replication":"https://pith.science/pith/POHEBDH5FJNAEVWOWO4UGGTQWG/action/replication_record"}},"created_at":"2026-07-05T10:29:22.529477+00:00","updated_at":"2026-07-05T10:29:22.529477+00:00"}