{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:3OGDTXPJI7RAK3HISIV75TJOEY","short_pith_number":"pith:3OGDTXPJ","schema_version":"1.0","canonical_sha256":"db8c39dde947e2056ce8922bfecd2e26109f6814af2c4c8b3edf53b3adbd58d3","source":{"kind":"arxiv","id":"2503.15875","version":1},"attestation_state":"computed","paper":{"title":"MiLA: Multi-view Intensive-fidelity Long-term Video Generation World Model for Autonomous Driving","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Bing Wang, Daqi Liu, Enhui Ma, Haiguang Wang, Haisong Liu, Hongwei Xie, Kaicheng Yu, Limin Wang","submitted_at":"2025-03-20T05:58:32Z","abstract_excerpt":"In recent years, data-driven techniques have greatly advanced autonomous driving systems, but the need for rare and diverse training data remains a challenge, requiring significant investment in equipment and labor. World models, which predict and generate future environmental states, offer a promising solution by synthesizing annotated video data for training. However, existing methods struggle to generate long, consistent videos without accumulating errors, especially in dynamic scenes. To address this, we propose MiLA, a novel framework for generating high-fidelity, long-duration videos up "},"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":"2503.15875","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-03-20T05:58:32Z","cross_cats_sorted":[],"title_canon_sha256":"b573556da9a3458219e92f95f66fe3eafd2ed0c57907447d3d1e3eb4c6adf22c","abstract_canon_sha256":"6d0e05001a18382750a208ce3002fdea0b9bb8d37ef15f129b4710c61c4c71b8"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:36:00.527895Z","signature_b64":"nED+1CPXL4xe+SBOj0ZHW1+KdGOehsLrpbkqS1/Okid+49KeCrD5r3MJCBZo/67NUaykp5ZLjGzzRiBXSGL6BQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"db8c39dde947e2056ce8922bfecd2e26109f6814af2c4c8b3edf53b3adbd58d3","last_reissued_at":"2026-07-05T10:36:00.526990Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:36:00.526990Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"MiLA: Multi-view Intensive-fidelity Long-term Video Generation World Model for Autonomous Driving","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Bing Wang, Daqi Liu, Enhui Ma, Haiguang Wang, Haisong Liu, Hongwei Xie, Kaicheng Yu, Limin Wang","submitted_at":"2025-03-20T05:58:32Z","abstract_excerpt":"In recent years, data-driven techniques have greatly advanced autonomous driving systems, but the need for rare and diverse training data remains a challenge, requiring significant investment in equipment and labor. World models, which predict and generate future environmental states, offer a promising solution by synthesizing annotated video data for training. However, existing methods struggle to generate long, consistent videos without accumulating errors, especially in dynamic scenes. To address this, we propose MiLA, a novel framework for generating high-fidelity, long-duration videos up "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2503.15875","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/2503.15875/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":"2503.15875","created_at":"2026-07-05T10:36:00.527116+00:00"},{"alias_kind":"arxiv_version","alias_value":"2503.15875v1","created_at":"2026-07-05T10:36:00.527116+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2503.15875","created_at":"2026-07-05T10:36:00.527116+00:00"},{"alias_kind":"pith_short_12","alias_value":"3OGDTXPJI7RA","created_at":"2026-07-05T10:36:00.527116+00:00"},{"alias_kind":"pith_short_16","alias_value":"3OGDTXPJI7RAK3HI","created_at":"2026-07-05T10:36:00.527116+00:00"},{"alias_kind":"pith_short_8","alias_value":"3OGDTXPJ","created_at":"2026-07-05T10:36:00.527116+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":5,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.17536","citing_title":"OmniDrive: An LLM-Choreographed Multi-Agent World Model with Unified Latent Co-Compression for Multi-View Driving Video Generation","ref_index":47,"is_internal_anchor":false},{"citing_arxiv_id":"2606.27504","citing_title":"ReWorld: Learning Better Representations for World Action Models","ref_index":26,"is_internal_anchor":false},{"citing_arxiv_id":"2606.29908","citing_title":"Pondering the Way: Spatial-perceiving World Action Model for Embodied Navigation","ref_index":40,"is_internal_anchor":false},{"citing_arxiv_id":"2512.23421","citing_title":"DriveLaW:Unifying Planning and Video Generation in a Latent Driving World","ref_index":65,"is_internal_anchor":false},{"citing_arxiv_id":"2603.28489","citing_title":"Video Generation Models as World Models: Efficient Paradigms, Architectures and Algorithms","ref_index":186,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/3OGDTXPJI7RAK3HISIV75TJOEY","json":"https://pith.science/pith/3OGDTXPJI7RAK3HISIV75TJOEY.json","graph_json":"https://pith.science/api/pith-number/3OGDTXPJI7RAK3HISIV75TJOEY/graph.json","events_json":"https://pith.science/api/pith-number/3OGDTXPJI7RAK3HISIV75TJOEY/events.json","paper":"https://pith.science/paper/3OGDTXPJ"},"agent_actions":{"view_html":"https://pith.science/pith/3OGDTXPJI7RAK3HISIV75TJOEY","download_json":"https://pith.science/pith/3OGDTXPJI7RAK3HISIV75TJOEY.json","view_paper":"https://pith.science/paper/3OGDTXPJ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2503.15875&json=true","fetch_graph":"https://pith.science/api/pith-number/3OGDTXPJI7RAK3HISIV75TJOEY/graph.json","fetch_events":"https://pith.science/api/pith-number/3OGDTXPJI7RAK3HISIV75TJOEY/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/3OGDTXPJI7RAK3HISIV75TJOEY/action/timestamp_anchor","attest_storage":"https://pith.science/pith/3OGDTXPJI7RAK3HISIV75TJOEY/action/storage_attestation","attest_author":"https://pith.science/pith/3OGDTXPJI7RAK3HISIV75TJOEY/action/author_attestation","sign_citation":"https://pith.science/pith/3OGDTXPJI7RAK3HISIV75TJOEY/action/citation_signature","submit_replication":"https://pith.science/pith/3OGDTXPJI7RAK3HISIV75TJOEY/action/replication_record"}},"created_at":"2026-07-05T10:36:00.527116+00:00","updated_at":"2026-07-05T10:36:00.527116+00:00"}