{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:LA2Q6IDSGP5ZTQHIACPBYKSNZK","short_pith_number":"pith:LA2Q6IDS","schema_version":"1.0","canonical_sha256":"58350f207233fb99c0e8009e1c2a4dca9c2f4336bbce73d17ff91f813e6d1ee4","source":{"kind":"arxiv","id":"2412.09262","version":2},"attestation_state":"computed","paper":{"title":"LatentSync: Taming Audio-Conditioned Latent Diffusion Models for Lip Sync with SyncNet Supervision","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Bingyue Peng, Chao Zhang, Chunyu Li, Cunjian Chen, Jinghui Xie, Jingyu Lin, Weiguo Feng, Weikai Xu, Weiwei Xing","submitted_at":"2024-12-12T13:20:52Z","abstract_excerpt":"End-to-end audio-conditioned latent diffusion models (LDMs) have been widely adopted for audio-driven portrait animation, demonstrating their effectiveness in generating lifelike and high-resolution talking videos. However, direct application of audio-conditioned LDMs to lip-synchronization (lip-sync) tasks results in suboptimal lip-sync accuracy. Through an in-depth analysis, we identified the underlying cause as the \"shortcut learning problem\", wherein the model predominantly learns visual-visual shortcuts while neglecting the critical audio-visual correlations. To address this issue, we exp"},"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":"2412.09262","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2024-12-12T13:20:52Z","cross_cats_sorted":[],"title_canon_sha256":"106ae65c9059d01004a8307cd378512a10d9bc3c50188c73930de4ff4282919e","abstract_canon_sha256":"a1c1711f5dada2232e2a8b3a727fa5ee853b896c3435dbb2639c7c7a2b71fb68"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:30:29.293422Z","signature_b64":"MLc5smuBePQy8x3DtxLnjqnNtvVpBYaCYo2KmqA1rpn0kdF8MfGPJ8kip80vsE2LgKOshiZsUGtauAWCjipWBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"58350f207233fb99c0e8009e1c2a4dca9c2f4336bbce73d17ff91f813e6d1ee4","last_reissued_at":"2026-07-05T10:30:29.292929Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:30:29.292929Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"LatentSync: Taming Audio-Conditioned Latent Diffusion Models for Lip Sync with SyncNet Supervision","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Bingyue Peng, Chao Zhang, Chunyu Li, Cunjian Chen, Jinghui Xie, Jingyu Lin, Weiguo Feng, Weikai Xu, Weiwei Xing","submitted_at":"2024-12-12T13:20:52Z","abstract_excerpt":"End-to-end audio-conditioned latent diffusion models (LDMs) have been widely adopted for audio-driven portrait animation, demonstrating their effectiveness in generating lifelike and high-resolution talking videos. However, direct application of audio-conditioned LDMs to lip-synchronization (lip-sync) tasks results in suboptimal lip-sync accuracy. Through an in-depth analysis, we identified the underlying cause as the \"shortcut learning problem\", wherein the model predominantly learns visual-visual shortcuts while neglecting the critical audio-visual correlations. To address this issue, we exp"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2412.09262","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/2412.09262/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":"2412.09262","created_at":"2026-07-05T10:30:29.292987+00:00"},{"alias_kind":"arxiv_version","alias_value":"2412.09262v2","created_at":"2026-07-05T10:30:29.292987+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2412.09262","created_at":"2026-07-05T10:30:29.292987+00:00"},{"alias_kind":"pith_short_12","alias_value":"LA2Q6IDSGP5Z","created_at":"2026-07-05T10:30:29.292987+00:00"},{"alias_kind":"pith_short_16","alias_value":"LA2Q6IDSGP5ZTQHI","created_at":"2026-07-05T10:30:29.292987+00:00"},{"alias_kind":"pith_short_8","alias_value":"LA2Q6IDS","created_at":"2026-07-05T10:30:29.292987+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":21,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.26092","citing_title":"TryOnCrafter: Unleashing Camera Trajectories for Realistic Video Virtual Try-on via a Renderable 4D Try-on Proxy","ref_index":19,"is_internal_anchor":false},{"citing_arxiv_id":"2606.22772","citing_title":"LoCC: Detection and Localization of Lip-Syncing Deepfakes via Counterfactual Frame Consistency","ref_index":16,"is_internal_anchor":false},{"citing_arxiv_id":"2606.11180","citing_title":"Lip Forcing: Few-Step Autoregressive Diffusion for Real-time Lip Synchronization","ref_index":25,"is_internal_anchor":false},{"citing_arxiv_id":"2606.03168","citing_title":"JAVEDIT: Joint Audio-Visual Instruction-Guided Video Editing with Agentic Data Curation","ref_index":28,"is_internal_anchor":false},{"citing_arxiv_id":"2605.20183","citing_title":"MSAVBench: Towards Comprehensive and Reliable Evaluation of Multi-Shot Audio-Video Generation","ref_index":32,"is_internal_anchor":false},{"citing_arxiv_id":"2605.24652","citing_title":"AVBench: Human-Aligned and Automated Evaluation Benchmark for Audio-Video Generative Models","ref_index":17,"is_internal_anchor":false},{"citing_arxiv_id":"2606.27779","citing_title":"MindFlow: Harmonizing Cognitive Semantics and Acoustic Dynamics for Facial Animation Generation in Dyadic Conversations","ref_index":26,"is_internal_anchor":false},{"citing_arxiv_id":"2605.28056","citing_title":"CogPortrait: Fine-Grained Eye-Region Control in Portrait Animation via Hierarchical Agent Planning","ref_index":30,"is_internal_anchor":false},{"citing_arxiv_id":"2605.30230","citing_title":"IP-Adapter Is All You Need: Towards Fine-Tuning-Free Diffusion-Based Talking Face Generation","ref_index":23,"is_internal_anchor":false},{"citing_arxiv_id":"2605.20183","citing_title":"MSAVBench: Towards Comprehensive and Reliable Evaluation of Multi-Shot Audio-Video Generation","ref_index":32,"is_internal_anchor":false},{"citing_arxiv_id":"2605.16918","citing_title":"HighSync: High-Quality Lip Synchronization via Latent Diffusion Models","ref_index":4,"is_internal_anchor":false},{"citing_arxiv_id":"2509.12052","citing_title":"FluentAvatar: Flicker-Free Talking-Head Animation via Phoneme-Guided Autoregressive Modeling","ref_index":15,"is_internal_anchor":false},{"citing_arxiv_id":"2512.09299","citing_title":"VABench: A Comprehensive Benchmark for Audio-Video Generation","ref_index":22,"is_internal_anchor":false},{"citing_arxiv_id":"2601.22143","citing_title":"JUST-DUB-IT: Video Dubbing via Joint Audio-Visual Diffusion","ref_index":12,"is_internal_anchor":false},{"citing_arxiv_id":"2602.09534","citing_title":"AUHead: Realistic Emotional Talking Head Generation via Action Units Control","ref_index":9,"is_internal_anchor":false},{"citing_arxiv_id":"2605.10334","citing_title":"The Alpha Blending Hypothesis: Compositing Shortcut in Deepfake Detection","ref_index":22,"is_internal_anchor":false},{"citing_arxiv_id":"2604.23632","citing_title":"Hallo-Live: Real-Time Streaming Joint Audio-Video Avatar Generation with Asynchronous Dual-Stream and Human-Centric Preference Distillation","ref_index":18,"is_internal_anchor":false},{"citing_arxiv_id":"2604.11804","citing_title":"OmniShow: Unifying Multimodal Conditions for Human-Object Interaction Video Generation","ref_index":35,"is_internal_anchor":false},{"citing_arxiv_id":"2604.10542","citing_title":"VidAudio-Bench: Benchmarking V2A and VT2A Generation across Four Audio Categories","ref_index":29,"is_internal_anchor":false},{"citing_arxiv_id":"2604.11283","citing_title":"Multimodal Large Language Model-Enabled Video Translation: A Role-Oriented Survey","ref_index":139,"is_internal_anchor":false},{"citing_arxiv_id":"2604.08405","citing_title":"SyncBreaker:Stage-Aware Multimodal Adversarial Attacks on Audio-Driven Talking Head Generation","ref_index":30,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/LA2Q6IDSGP5ZTQHIACPBYKSNZK","json":"https://pith.science/pith/LA2Q6IDSGP5ZTQHIACPBYKSNZK.json","graph_json":"https://pith.science/api/pith-number/LA2Q6IDSGP5ZTQHIACPBYKSNZK/graph.json","events_json":"https://pith.science/api/pith-number/LA2Q6IDSGP5ZTQHIACPBYKSNZK/events.json","paper":"https://pith.science/paper/LA2Q6IDS"},"agent_actions":{"view_html":"https://pith.science/pith/LA2Q6IDSGP5ZTQHIACPBYKSNZK","download_json":"https://pith.science/pith/LA2Q6IDSGP5ZTQHIACPBYKSNZK.json","view_paper":"https://pith.science/paper/LA2Q6IDS","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2412.09262&json=true","fetch_graph":"https://pith.science/api/pith-number/LA2Q6IDSGP5ZTQHIACPBYKSNZK/graph.json","fetch_events":"https://pith.science/api/pith-number/LA2Q6IDSGP5ZTQHIACPBYKSNZK/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/LA2Q6IDSGP5ZTQHIACPBYKSNZK/action/timestamp_anchor","attest_storage":"https://pith.science/pith/LA2Q6IDSGP5ZTQHIACPBYKSNZK/action/storage_attestation","attest_author":"https://pith.science/pith/LA2Q6IDSGP5ZTQHIACPBYKSNZK/action/author_attestation","sign_citation":"https://pith.science/pith/LA2Q6IDSGP5ZTQHIACPBYKSNZK/action/citation_signature","submit_replication":"https://pith.science/pith/LA2Q6IDSGP5ZTQHIACPBYKSNZK/action/replication_record"}},"created_at":"2026-07-05T10:30:29.292987+00:00","updated_at":"2026-07-05T10:30:29.292987+00:00"}