{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:7NLFFWOBUFPWWKQRD4NXRATHGW","short_pith_number":"pith:7NLFFWOB","schema_version":"1.0","canonical_sha256":"fb5652d9c1a15f6b2a111f1b788267358946b746507846aac867f099aa1f0551","source":{"kind":"arxiv","id":"2409.12514","version":5},"attestation_state":"computed","paper":{"title":"TinyVLA: Towards Fast, Data-Efficient Vision-Language-Action Models for Robotic Manipulation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"TinyVLA reaches OpenVLA-level performance on robot tasks by initializing from fast multimodal models and adding a diffusion action decoder, removing the pre-training stage entirely.","cross_cats":["cs.CV"],"primary_cat":"cs.RO","authors_text":"Chaomin Shen, Feifei Feng, Jian Tang, Jinming Li, Junjie Wen, Kun Wu, Minjie Zhu, Ning Liu, Ran Cheng, Yaxin Peng, Yichen Zhu, Zhiyuan Xu","submitted_at":"2024-09-19T07:10:18Z","abstract_excerpt":"Vision-Language-Action (VLA) models have shown remarkable potential in visuomotor control and instruction comprehension through end-to-end learning processes. However, current VLA models face significant challenges: they are slow during inference and require extensive pre-training on large amounts of robotic data, making real-world deployment difficult. In this paper, we introduce a new family of compact vision-language-action models, called TinyVLA, which offers two key advantages over existing VLA models: (1) faster inference speeds, and (2) improved data efficiency, eliminating the need for"},"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":true,"formal_links_present":true},"canonical_record":{"source":{"id":"2409.12514","kind":"arxiv","version":5},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.RO","submitted_at":"2024-09-19T07:10:18Z","cross_cats_sorted":["cs.CV"],"title_canon_sha256":"29fb64fb48e6a5eb3ad63b2ee904bee527a38ffc0b152de0baa1808b18c63dc6","abstract_canon_sha256":"4b8dc1f54b57add6d3130db258de0181fe6b46834717ccd6fdea4538564fde06"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:38:13.622548Z","signature_b64":"7jpM5vjfNW6XtMEbWjAbSBYApo2J1+Ukq+ogol73XJA37+FJthyG4WkMnamjn7GGIxSfbGql3Wv5dvnDIesiBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"fb5652d9c1a15f6b2a111f1b788267358946b746507846aac867f099aa1f0551","last_reissued_at":"2026-05-17T23:38:13.621854Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:38:13.621854Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"TinyVLA: Towards Fast, Data-Efficient Vision-Language-Action Models for Robotic Manipulation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"TinyVLA reaches OpenVLA-level performance on robot tasks by initializing from fast multimodal models and adding a diffusion action decoder, removing the pre-training stage entirely.","cross_cats":["cs.CV"],"primary_cat":"cs.RO","authors_text":"Chaomin Shen, Feifei Feng, Jian Tang, Jinming Li, Junjie Wen, Kun Wu, Minjie Zhu, Ning Liu, Ran Cheng, Yaxin Peng, Yichen Zhu, Zhiyuan Xu","submitted_at":"2024-09-19T07:10:18Z","abstract_excerpt":"Vision-Language-Action (VLA) models have shown remarkable potential in visuomotor control and instruction comprehension through end-to-end learning processes. However, current VLA models face significant challenges: they are slow during inference and require extensive pre-training on large amounts of robotic data, making real-world deployment difficult. In this paper, we introduce a new family of compact vision-language-action models, called TinyVLA, which offers two key advantages over existing VLA models: (1) faster inference speeds, and (2) improved data efficiency, eliminating the need for"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"Our approach significantly outperforms the state-of-the-art VLA model, OpenVLA, in terms of speed and data efficiency, while delivering comparable or superior performance.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That initializing the policy backbone with existing high-speed multimodal models plus a diffusion decoder during fine-tuning is sufficient to eliminate the pre-training stage while preserving or improving task performance and generalization.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"TinyVLA achieves faster inference and higher data efficiency than OpenVLA on robotic manipulation tasks by initializing from high-speed multimodal models and adding a diffusion policy decoder, without any pre-training phase.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"TinyVLA reaches OpenVLA-level performance on robot tasks by initializing from fast multimodal models and adding a diffusion action decoder, removing the pre-training stage entirely.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"d2290bf4b1530764ac0538952466e4728758aebbbdb0d343407da3c0cedc10b3"},"source":{"id":"2409.12514","kind":"arxiv","version":5},"verdict":{"id":"b4940e26-1ddf-4bd7-8052-f922074b9249","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-17T16:07:08.380660Z","strongest_claim":"Our approach significantly outperforms the state-of-the-art VLA model, OpenVLA, in terms of speed and data efficiency, while delivering comparable or superior performance.","one_line_summary":"TinyVLA achieves faster inference and higher data efficiency than OpenVLA on robotic manipulation tasks by initializing from high-speed multimodal models and adding a diffusion policy decoder, without any pre-training phase.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That initializing the policy backbone with existing high-speed multimodal models plus a diffusion decoder during fine-tuning is sufficient to eliminate the pre-training stage while preserving or improving task performance and generalization.","pith_extraction_headline":"TinyVLA reaches OpenVLA-level performance on robot tasks by initializing from fast multimodal models and adding a diffusion action decoder, removing the pre-training stage entirely."},"references":{"count":46,"sample":[{"doi":"","year":2024,"title":"Roboagent: Generalization and efficiency in robot manipulation via semantic augmentations and action chunking,","work_id":"ad2bd81c-bdcd-4b21-bc80-5819aaa4bea7","ref_index":1,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2022,"title":"Bridge data: Boosting generalization of robotic skills with cross-domain datasets,","work_id":"fdece732-d4e5-450c-8f75-6ca0dbdf84ac","ref_index":2,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2023,"title":"Diffusion policy: Visuomotor policy learning via action diffusion,","work_id":"eaca3997-9409-4ca0-a030-e58e1eae24f6","ref_index":3,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2024,"title":"3d diffusion policy: Generalizable visuomo- tor policy learning via simple 3d representations,","work_id":"709db3be-7760-4bb5-bad0-1fc5c10cfd2f","ref_index":4,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2023,"title":"Llama 2: Open Foundation and Fine-Tuned Chat Models","work_id":"68a5177f-d644-44c1-bd4f-4e5278c22f5d","ref_index":5,"cited_arxiv_id":"2307.09288","is_internal_anchor":true}],"resolved_work":46,"snapshot_sha256":"40a7bc759fc5fcf4399418d23e48a0662db7903bbb5d67e27a2eacbe096ef5bc","internal_anchors":11},"formal_canon":{"evidence_count":1,"snapshot_sha256":"03095fa8158cd570a3a929796d717e55d443cc61cc3ce48bb56e3987742f6747"},"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":"2409.12514","created_at":"2026-05-17T23:38:13.621960+00:00"},{"alias_kind":"arxiv_version","alias_value":"2409.12514v5","created_at":"2026-05-17T23:38:13.621960+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2409.12514","created_at":"2026-05-17T23:38:13.621960+00:00"},{"alias_kind":"pith_short_12","alias_value":"7NLFFWOBUFPW","created_at":"2026-05-18T12:33:37.589309+00:00"},{"alias_kind":"pith_short_16","alias_value":"7NLFFWOBUFPWWKQR","created_at":"2026-05-18T12:33:37.589309+00:00"},{"alias_kind":"pith_short_8","alias_value":"7NLFFWOB","created_at":"2026-05-18T12:33:37.589309+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":24,"internal_anchor_count":24,"sample":[{"citing_arxiv_id":"2405.14093","citing_title":"A Survey on Vision-Language-Action Models for Embodied AI","ref_index":123,"is_internal_anchor":true},{"citing_arxiv_id":"2504.16054","citing_title":"$\\pi_{0.5}$: a Vision-Language-Action Model with Open-World Generalization","ref_index":83,"is_internal_anchor":true},{"citing_arxiv_id":"2505.03500","citing_title":"VLAs are Confined yet Capable of Generalizing to Novel Instructions","ref_index":42,"is_internal_anchor":true},{"citing_arxiv_id":"2605.21854","citing_title":"CrossVLA: Cross-Paradigm Post-Training and Inference Optimization for Vision-Language-Action Models","ref_index":13,"is_internal_anchor":true},{"citing_arxiv_id":"2605.16241","citing_title":"Offline Semantic Guidance for Efficient Vision-Language-Action Policy Distillation","ref_index":17,"is_internal_anchor":true},{"citing_arxiv_id":"2605.17486","citing_title":"DyGRO-VLA: Cross-Task Scaling of Vision-Language-Action Models via Dynamic Grouped Residual Optimization","ref_index":47,"is_internal_anchor":true},{"citing_arxiv_id":"2505.03233","citing_title":"GraspVLA: a Grasping Foundation Model Pre-trained on Billion-scale Synthetic Action Data","ref_index":28,"is_internal_anchor":true},{"citing_arxiv_id":"2505.15659","citing_title":"FLARE: Robot Learning with Implicit World Modeling","ref_index":36,"is_internal_anchor":true},{"citing_arxiv_id":"2503.22020","citing_title":"CoT-VLA: Visual Chain-of-Thought Reasoning for Vision-Language-Action Models","ref_index":65,"is_internal_anchor":true},{"citing_arxiv_id":"2502.19417","citing_title":"Hi Robot: Open-Ended Instruction Following with Hierarchical Vision-Language-Action Models","ref_index":46,"is_internal_anchor":true},{"citing_arxiv_id":"2503.10631","citing_title":"HybridVLA: Collaborative Diffusion and Autoregression in a Unified Vision-Language-Action Model","ref_index":61,"is_internal_anchor":true},{"citing_arxiv_id":"2502.05855","citing_title":"DexVLA: Vision-Language Model with Plug-In Diffusion Expert for General Robot Control","ref_index":43,"is_internal_anchor":true},{"citing_arxiv_id":"2504.02792","citing_title":"Unified World Models: Coupling Video and Action Diffusion for Pretraining on Large Robotic Datasets","ref_index":45,"is_internal_anchor":true},{"citing_arxiv_id":"2605.11665","citing_title":"Nautilus: From One Prompt to Plug-and-Play Robot Learning","ref_index":54,"is_internal_anchor":true},{"citing_arxiv_id":"2411.19650","citing_title":"CogACT: A Foundational Vision-Language-Action Model for Synergizing Cognition and Action in Robotic Manipulation","ref_index":67,"is_internal_anchor":true},{"citing_arxiv_id":"2605.10925","citing_title":"PriorVLA: Prior-Preserving Adaptation for Vision-Language-Action Models","ref_index":15,"is_internal_anchor":true},{"citing_arxiv_id":"2506.01844","citing_title":"SmolVLA: A Vision-Language-Action Model for Affordable and Efficient Robotics","ref_index":44,"is_internal_anchor":true},{"citing_arxiv_id":"2501.09747","citing_title":"FAST: Efficient Action Tokenization for Vision-Language-Action Models","ref_index":64,"is_internal_anchor":true},{"citing_arxiv_id":"2502.19645","citing_title":"Fine-Tuning Vision-Language-Action Models: Optimizing Speed and Success","ref_index":55,"is_internal_anchor":true},{"citing_arxiv_id":"2604.05672","citing_title":"A1: A Fully Transparent Open-Source, Adaptive and Efficient Truncated Vision-Language-Action Model","ref_index":37,"is_internal_anchor":true},{"citing_arxiv_id":"2503.14734","citing_title":"GR00T N1: An Open Foundation Model for Generalist Humanoid Robots","ref_index":91,"is_internal_anchor":true},{"citing_arxiv_id":"2410.24164","citing_title":"$\\pi_0$: A Vision-Language-Action Flow Model for General Robot Control","ref_index":55,"is_internal_anchor":true},{"citing_arxiv_id":"2604.15483","citing_title":"${\\pi}_{0.7}$: a Steerable Generalist Robotic Foundation Model with Emergent Capabilities","ref_index":11,"is_internal_anchor":true},{"citing_arxiv_id":"2604.17800","citing_title":"ReFineVLA: Multimodal Reasoning-Aware Generalist Robotic Policies via Teacher-Guided Fine-Tuning","ref_index":40,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":1,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/7NLFFWOBUFPWWKQRD4NXRATHGW","json":"https://pith.science/pith/7NLFFWOBUFPWWKQRD4NXRATHGW.json","graph_json":"https://pith.science/api/pith-number/7NLFFWOBUFPWWKQRD4NXRATHGW/graph.json","events_json":"https://pith.science/api/pith-number/7NLFFWOBUFPWWKQRD4NXRATHGW/events.json","paper":"https://pith.science/paper/7NLFFWOB"},"agent_actions":{"view_html":"https://pith.science/pith/7NLFFWOBUFPWWKQRD4NXRATHGW","download_json":"https://pith.science/pith/7NLFFWOBUFPWWKQRD4NXRATHGW.json","view_paper":"https://pith.science/paper/7NLFFWOB","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2409.12514&json=true","fetch_graph":"https://pith.science/api/pith-number/7NLFFWOBUFPWWKQRD4NXRATHGW/graph.json","fetch_events":"https://pith.science/api/pith-number/7NLFFWOBUFPWWKQRD4NXRATHGW/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/7NLFFWOBUFPWWKQRD4NXRATHGW/action/timestamp_anchor","attest_storage":"https://pith.science/pith/7NLFFWOBUFPWWKQRD4NXRATHGW/action/storage_attestation","attest_author":"https://pith.science/pith/7NLFFWOBUFPWWKQRD4NXRATHGW/action/author_attestation","sign_citation":"https://pith.science/pith/7NLFFWOBUFPWWKQRD4NXRATHGW/action/citation_signature","submit_replication":"https://pith.science/pith/7NLFFWOBUFPWWKQRD4NXRATHGW/action/replication_record"}},"created_at":"2026-05-17T23:38:13.621960+00:00","updated_at":"2026-05-17T23:38:13.621960+00:00"}