{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:QOSUERWVEOH3BK3V7PM7CBG24K","short_pith_number":"pith:QOSUERWV","schema_version":"1.0","canonical_sha256":"83a54246d5238fb0ab75fbd9f104dae2a2a4dda6717c6199a520bdc478d2a817","source":{"kind":"arxiv","id":"2411.17335","version":2},"attestation_state":"computed","paper":{"title":"VersatileMotion: A Unified Framework for Motion Synthesis and Comprehension","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Bo Han, Changqing Zou, Hongdeng Shen, Jikang Cheng, Shiyang Li, Zeyu Ling","submitted_at":"2024-11-26T11:28:01Z","abstract_excerpt":"Large language models (LLMs) are, by design, inherently capable of multi-task learning: through a unified next-token prediction paradigm, they can naturally address a wide variety of downstream tasks. Prior work in the motion domain has demonstrated some generality by adapting LLMs via a Motion Tokenizer coupled with an autoregressive Transformer to generate and understand human motion. However, this generality remains limited in scope and yields only modest performance gains. We introduce VersatileMotion, a unified multimodal motion LLM that combines a novel motion tokenizer, integrating VQ-V"},"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":"2411.17335","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2024-11-26T11:28:01Z","cross_cats_sorted":[],"title_canon_sha256":"7e8bb53a946ce2cf5e8cbe7f89c6de3680bc19571b30fbc27f31a4576c801f18","abstract_canon_sha256":"766a4492978df8a300c1e84faa0f4d171eeba48b7c40f12bc9dc88fd09245f6a"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:09:44.128413Z","signature_b64":"AfnxEWWQkoQGg+MooaTslGeFtaQs/9rILOe7G8LBQg+rVnmUmdMMPPqeeq8CfUfP26vbStY9SMe9OksYwz6nBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"83a54246d5238fb0ab75fbd9f104dae2a2a4dda6717c6199a520bdc478d2a817","last_reissued_at":"2026-07-05T11:09:44.127923Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:09:44.127923Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"VersatileMotion: A Unified Framework for Motion Synthesis and Comprehension","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Bo Han, Changqing Zou, Hongdeng Shen, Jikang Cheng, Shiyang Li, Zeyu Ling","submitted_at":"2024-11-26T11:28:01Z","abstract_excerpt":"Large language models (LLMs) are, by design, inherently capable of multi-task learning: through a unified next-token prediction paradigm, they can naturally address a wide variety of downstream tasks. Prior work in the motion domain has demonstrated some generality by adapting LLMs via a Motion Tokenizer coupled with an autoregressive Transformer to generate and understand human motion. However, this generality remains limited in scope and yields only modest performance gains. We introduce VersatileMotion, a unified multimodal motion LLM that combines a novel motion tokenizer, integrating VQ-V"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2411.17335","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/2411.17335/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":"2411.17335","created_at":"2026-07-05T11:09:44.127979+00:00"},{"alias_kind":"arxiv_version","alias_value":"2411.17335v2","created_at":"2026-07-05T11:09:44.127979+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2411.17335","created_at":"2026-07-05T11:09:44.127979+00:00"},{"alias_kind":"pith_short_12","alias_value":"QOSUERWVEOH3","created_at":"2026-07-05T11:09:44.127979+00:00"},{"alias_kind":"pith_short_16","alias_value":"QOSUERWVEOH3BK3V","created_at":"2026-07-05T11:09:44.127979+00:00"},{"alias_kind":"pith_short_8","alias_value":"QOSUERWV","created_at":"2026-07-05T11:09:44.127979+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":5,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.21135","citing_title":"Odoriko: A Shape-Aware Multimodal Diffusion Framework for Human Motion","ref_index":37,"is_internal_anchor":false},{"citing_arxiv_id":"2606.04773","citing_title":"NextMotionQA: Benchmarking and Judging Human Motion Understanding with Vision-Language Models","ref_index":26,"is_internal_anchor":false},{"citing_arxiv_id":"2605.14731","citing_title":"UMo: Unified Sparse Motion Modeling for Real-Time Co-Speech Avatars","ref_index":31,"is_internal_anchor":false},{"citing_arxiv_id":"2605.29906","citing_title":"Plan, Don't Pose: Long Composite Motion Generation with Text-Aligned BFM","ref_index":16,"is_internal_anchor":false},{"citing_arxiv_id":"2605.29488","citing_title":"AnyMo: Scaling Any-Modality Conditional Motion Generation with Masked Modeling","ref_index":23,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/QOSUERWVEOH3BK3V7PM7CBG24K","json":"https://pith.science/pith/QOSUERWVEOH3BK3V7PM7CBG24K.json","graph_json":"https://pith.science/api/pith-number/QOSUERWVEOH3BK3V7PM7CBG24K/graph.json","events_json":"https://pith.science/api/pith-number/QOSUERWVEOH3BK3V7PM7CBG24K/events.json","paper":"https://pith.science/paper/QOSUERWV"},"agent_actions":{"view_html":"https://pith.science/pith/QOSUERWVEOH3BK3V7PM7CBG24K","download_json":"https://pith.science/pith/QOSUERWVEOH3BK3V7PM7CBG24K.json","view_paper":"https://pith.science/paper/QOSUERWV","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2411.17335&json=true","fetch_graph":"https://pith.science/api/pith-number/QOSUERWVEOH3BK3V7PM7CBG24K/graph.json","fetch_events":"https://pith.science/api/pith-number/QOSUERWVEOH3BK3V7PM7CBG24K/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/QOSUERWVEOH3BK3V7PM7CBG24K/action/timestamp_anchor","attest_storage":"https://pith.science/pith/QOSUERWVEOH3BK3V7PM7CBG24K/action/storage_attestation","attest_author":"https://pith.science/pith/QOSUERWVEOH3BK3V7PM7CBG24K/action/author_attestation","sign_citation":"https://pith.science/pith/QOSUERWVEOH3BK3V7PM7CBG24K/action/citation_signature","submit_replication":"https://pith.science/pith/QOSUERWVEOH3BK3V7PM7CBG24K/action/replication_record"}},"created_at":"2026-07-05T11:09:44.127979+00:00","updated_at":"2026-07-05T11:09:44.127979+00:00"}