{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:DFNIMTTEWTSRMD7V2V556MW2CB","short_pith_number":"pith:DFNIMTTE","schema_version":"1.0","canonical_sha256":"195a864e64b4e5160ff5d57bdf32da10786ce5489ed636ded19d94b400985dd6","source":{"kind":"arxiv","id":"2503.20491","version":2},"attestation_state":"computed","paper":{"title":"VPO: Aligning Text-to-Video Generation Models with Prompt Optimization","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL","cs.LG"],"primary_cat":"cs.CV","authors_text":"Hongning Wang, Jiale Cheng, Jiayan Teng, Jiazheng Xu, Jie Tang, Minlie Huang, Ruiliang Lyu, Xiao Liu, Xiaotao Gu, Yida Lu, Yuxiao Dong, Zhuoyi Yang","submitted_at":"2025-03-26T12:28:20Z","abstract_excerpt":"Video generation models have achieved remarkable progress in text-to-video tasks. These models are typically trained on text-video pairs with highly detailed and carefully crafted descriptions, while real-world user inputs during inference are often concise, vague, or poorly structured. This gap makes prompt optimization crucial for generating high-quality videos. Current methods often rely on large language models (LLMs) to refine prompts through in-context learning, but suffer from several limitations: they may distort user intent, omit critical details, or introduce safety risks. Moreover, "},"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":"2503.20491","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-03-26T12:28:20Z","cross_cats_sorted":["cs.CL","cs.LG"],"title_canon_sha256":"75b3c90d36548045bbb7be0e2c34d8ccb6070eabc8345449c206c4c02ff37d05","abstract_canon_sha256":"2cb148e9bfcc4817eff2c11f34347c96622888e56f0ef360f5ec9cc000c3f33f"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T12:01:59.455258Z","signature_b64":"l6bnSBQfGADAeUnL8E5xGP6Ht8wVarSgP5zP64oGVFvJFjFe2hsHxfZ1t2LrDDfnQx+HilhtrXkyPKc3SBZUCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"195a864e64b4e5160ff5d57bdf32da10786ce5489ed636ded19d94b400985dd6","last_reissued_at":"2026-07-05T12:01:59.454660Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T12:01:59.454660Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"VPO: Aligning Text-to-Video Generation Models with Prompt Optimization","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL","cs.LG"],"primary_cat":"cs.CV","authors_text":"Hongning Wang, Jiale Cheng, Jiayan Teng, Jiazheng Xu, Jie Tang, Minlie Huang, Ruiliang Lyu, Xiao Liu, Xiaotao Gu, Yida Lu, Yuxiao Dong, Zhuoyi Yang","submitted_at":"2025-03-26T12:28:20Z","abstract_excerpt":"Video generation models have achieved remarkable progress in text-to-video tasks. These models are typically trained on text-video pairs with highly detailed and carefully crafted descriptions, while real-world user inputs during inference are often concise, vague, or poorly structured. This gap makes prompt optimization crucial for generating high-quality videos. Current methods often rely on large language models (LLMs) to refine prompts through in-context learning, but suffer from several limitations: they may distort user intent, omit critical details, or introduce safety risks. Moreover, "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2503.20491","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/2503.20491/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":"2503.20491","created_at":"2026-07-05T12:01:59.454737+00:00"},{"alias_kind":"arxiv_version","alias_value":"2503.20491v2","created_at":"2026-07-05T12:01:59.454737+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2503.20491","created_at":"2026-07-05T12:01:59.454737+00:00"},{"alias_kind":"pith_short_12","alias_value":"DFNIMTTEWTSR","created_at":"2026-07-05T12:01:59.454737+00:00"},{"alias_kind":"pith_short_16","alias_value":"DFNIMTTEWTSRMD7V","created_at":"2026-07-05T12:01:59.454737+00:00"},{"alias_kind":"pith_short_8","alias_value":"DFNIMTTE","created_at":"2026-07-05T12:01:59.454737+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.18591","citing_title":"Bridging Creative Intent and Visual Quality: Creator-Driven Recurrent Video Generation with Agentic Feedback Loops","ref_index":2,"is_internal_anchor":false},{"citing_arxiv_id":"2512.04678","citing_title":"Reward Forcing: Efficient Streaming Video Generation with Rewarded Distribution Matching Distillation","ref_index":10,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/DFNIMTTEWTSRMD7V2V556MW2CB","json":"https://pith.science/pith/DFNIMTTEWTSRMD7V2V556MW2CB.json","graph_json":"https://pith.science/api/pith-number/DFNIMTTEWTSRMD7V2V556MW2CB/graph.json","events_json":"https://pith.science/api/pith-number/DFNIMTTEWTSRMD7V2V556MW2CB/events.json","paper":"https://pith.science/paper/DFNIMTTE"},"agent_actions":{"view_html":"https://pith.science/pith/DFNIMTTEWTSRMD7V2V556MW2CB","download_json":"https://pith.science/pith/DFNIMTTEWTSRMD7V2V556MW2CB.json","view_paper":"https://pith.science/paper/DFNIMTTE","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2503.20491&json=true","fetch_graph":"https://pith.science/api/pith-number/DFNIMTTEWTSRMD7V2V556MW2CB/graph.json","fetch_events":"https://pith.science/api/pith-number/DFNIMTTEWTSRMD7V2V556MW2CB/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/DFNIMTTEWTSRMD7V2V556MW2CB/action/timestamp_anchor","attest_storage":"https://pith.science/pith/DFNIMTTEWTSRMD7V2V556MW2CB/action/storage_attestation","attest_author":"https://pith.science/pith/DFNIMTTEWTSRMD7V2V556MW2CB/action/author_attestation","sign_citation":"https://pith.science/pith/DFNIMTTEWTSRMD7V2V556MW2CB/action/citation_signature","submit_replication":"https://pith.science/pith/DFNIMTTEWTSRMD7V2V556MW2CB/action/replication_record"}},"created_at":"2026-07-05T12:01:59.454737+00:00","updated_at":"2026-07-05T12:01:59.454737+00:00"}