{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:6Z5RRGFM332ONHQZVDEEGS642O","short_pith_number":"pith:6Z5RRGFM","schema_version":"1.0","canonical_sha256":"f67b1898acdef4e69e19a8c8434bdcd39e9e70b9d9b5a0221c04811458bb201c","source":{"kind":"arxiv","id":"2505.03233","version":3},"attestation_state":"computed","paper":{"title":"GraspVLA: a Grasping Foundation Model Pre-trained on Billion-scale Synthetic Action Data","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"A grasping model pretrained entirely on a billion-frame synthetic dataset achieves open-vocabulary generalization to real robots by unifying perception and action in one chain-of-thought sequence.","cross_cats":[],"primary_cat":"cs.RO","authors_text":"Haixin Ma, Heming Cui, He Wang, Jiayi Chen, Mi Yan, Shengliang Deng, Songlin Wei, Taoyu Yang, Wenhao Zhang, Xuheng Zhang, Yuxin Yang, Zhiqi Zhang, Zhizheng Zhang","submitted_at":"2025-05-06T06:59:28Z","abstract_excerpt":"Embodied foundation models are gaining increasing attention for their zero-shot generalization, scalability, and adaptability to new tasks through few-shot post-training. However, existing models rely heavily on real-world data, which is costly and labor-intensive to collect. Synthetic data offers a cost-effective alternative, yet its potential remains largely underexplored. To bridge this gap, we explore the feasibility of training Vision-Language-Action models entirely with large-scale synthetic action data. We curate SynGrasp-1B, a billion-frame robotic grasping dataset generated in simulat"},"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":"2505.03233","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.RO","submitted_at":"2025-05-06T06:59:28Z","cross_cats_sorted":[],"title_canon_sha256":"46ca4618ee18801ced1bd632a133d2bc867615cc1840e7dd4d74851826e99ee7","abstract_canon_sha256":"efbbdfc7a28e25ef1c855fe6587059c150e48de3c8a7e0fd55e5e4426f7fa66e"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:38:13.103225Z","signature_b64":"UudoqpFAFVxwENMhXSjhM5G3NNNs7lONHnKtXitR6xugjbBBfZLV4sgGV0YSQDe+z235ipfM+XCLv67zVLUpBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f67b1898acdef4e69e19a8c8434bdcd39e9e70b9d9b5a0221c04811458bb201c","last_reissued_at":"2026-05-17T23:38:13.102573Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:38:13.102573Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"GraspVLA: a Grasping Foundation Model Pre-trained on Billion-scale Synthetic Action Data","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"A grasping model pretrained entirely on a billion-frame synthetic dataset achieves open-vocabulary generalization to real robots by unifying perception and action in one chain-of-thought sequence.","cross_cats":[],"primary_cat":"cs.RO","authors_text":"Haixin Ma, Heming Cui, He Wang, Jiayi Chen, Mi Yan, Shengliang Deng, Songlin Wei, Taoyu Yang, Wenhao Zhang, Xuheng Zhang, Yuxin Yang, Zhiqi Zhang, Zhizheng Zhang","submitted_at":"2025-05-06T06:59:28Z","abstract_excerpt":"Embodied foundation models are gaining increasing attention for their zero-shot generalization, scalability, and adaptability to new tasks through few-shot post-training. However, existing models rely heavily on real-world data, which is costly and labor-intensive to collect. Synthetic data offers a cost-effective alternative, yet its potential remains largely underexplored. To bridge this gap, we explore the feasibility of training Vision-Language-Action models entirely with large-scale synthetic action data. We curate SynGrasp-1B, a billion-frame robotic grasping dataset generated in simulat"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"GraspVLA integrates autoregressive perception tasks and flow-matching-based action generation into a unified Chain-of-Thought process, enabling joint training on synthetic action data and Internet semantics data to achieve open-vocabulary generalization in grasping.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That photorealistic rendering plus extensive domain randomization in simulation, combined with the CoT architecture, is sufficient to close the sim-to-real gap so that actions learned on synthetic data transfer effectively to physical robots on unseen objects.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"GraspVLA shows that pretraining a grasping model on a billion synthetic action frames enables zero-shot open-vocabulary performance and sim-to-real transfer.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"A grasping model pretrained entirely on a billion-frame synthetic dataset achieves open-vocabulary generalization to real robots by unifying perception and action in one chain-of-thought sequence.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"188944180b282cc6da36a10847252afadf58f39f68815391e810cf3bdbe88d8d"},"source":{"id":"2505.03233","kind":"arxiv","version":3},"verdict":{"id":"9dec9f6d-eea2-43b3-aec4-9d4bdd3e3b29","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-17T20:51:38.996102Z","strongest_claim":"GraspVLA integrates autoregressive perception tasks and flow-matching-based action generation into a unified Chain-of-Thought process, enabling joint training on synthetic action data and Internet semantics data to achieve open-vocabulary generalization in grasping.","one_line_summary":"GraspVLA shows that pretraining a grasping model on a billion synthetic action frames enables zero-shot open-vocabulary performance and sim-to-real transfer.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That photorealistic rendering plus extensive domain randomization in simulation, combined with the CoT architecture, is sufficient to close the sim-to-real gap so that actions learned on synthetic data transfer effectively to physical robots on unseen objects.","pith_extraction_headline":"A grasping model pretrained entirely on a billion-frame synthetic dataset achieves open-vocabulary generalization to real robots by unifying perception and action in one chain-of-thought sequence."},"references":{"count":90,"sample":[{"doi":"","year":2023,"title":"LLaMA: Open and Efficient Foundation Language Models","work_id":"c018fc23-6f3f-4035-9d02-28a2173b2b9d","ref_index":1,"cited_arxiv_id":"2302.13971","is_internal_anchor":true},{"doi":"","year":2023,"title":"Segment Anything","work_id":"2bbf46ca-720a-45a1-8e9c-10c33fbeada0","ref_index":2,"cited_arxiv_id":"2304.02643","is_internal_anchor":true},{"doi":"","year":2021,"title":"Learning Transferable Visual Models From Natural Language Supervision","work_id":"6de86bb5-27bd-4d5c-8b89-967ebfc52659","ref_index":3,"cited_arxiv_id":"2103.00020","is_internal_anchor":true},{"doi":"","year":2023,"title":"Chatgpt: Jan 17 version","work_id":"e752cccc-4880-4947-bff8-1b4c3e4bb107","ref_index":4,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2023,"title":"RT-2: Vision-Language-Action Models Transfer Web Knowledge to Robotic Control","work_id":"ff438a8a-8003-4fae-9131-acd418b3597b","ref_index":5,"cited_arxiv_id":"2307.15818","is_internal_anchor":true}],"resolved_work":90,"snapshot_sha256":"5cb778e1a6f55163ff9301275a3d0b10174854280aee400a6581d29cf1ba97af","internal_anchors":35},"formal_canon":{"evidence_count":1,"snapshot_sha256":"44a23072815f53252a63d673ee7d458ebdc8aa6fc2aa549dcd376c9f1dd6c8e2"},"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":"2505.03233","created_at":"2026-05-17T23:38:13.102696+00:00"},{"alias_kind":"arxiv_version","alias_value":"2505.03233v3","created_at":"2026-05-17T23:38:13.102696+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.03233","created_at":"2026-05-17T23:38:13.102696+00:00"},{"alias_kind":"pith_short_12","alias_value":"6Z5RRGFM332O","created_at":"2026-05-18T12:33:37.589309+00:00"},{"alias_kind":"pith_short_16","alias_value":"6Z5RRGFM332ONHQZ","created_at":"2026-05-18T12:33:37.589309+00:00"},{"alias_kind":"pith_short_8","alias_value":"6Z5RRGFM","created_at":"2026-05-18T12:33:37.589309+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":18,"internal_anchor_count":18,"sample":[{"citing_arxiv_id":"2508.13073","citing_title":"Large VLM-based Vision-Language-Action Models for Robotic Manipulation: A Survey","ref_index":143,"is_internal_anchor":true},{"citing_arxiv_id":"2601.02078","citing_title":"Genie Sim 3.0 : A High-Fidelity Comprehensive Simulation Platform for Humanoid Robot","ref_index":10,"is_internal_anchor":true},{"citing_arxiv_id":"2507.04447","citing_title":"DreamVLA: A Vision-Language-Action Model Dreamed with Comprehensive World Knowledge","ref_index":83,"is_internal_anchor":true},{"citing_arxiv_id":"2601.07060","citing_title":"PALM: Progress-Aware Policy Learning via Affordance Reasoning for Long-Horizon Robotic Manipulation","ref_index":21,"is_internal_anchor":true},{"citing_arxiv_id":"2601.16046","citing_title":"DextER: Language-driven Dexterous Grasp Generation with Embodied Reasoning","ref_index":9,"is_internal_anchor":true},{"citing_arxiv_id":"2605.13925","citing_title":"Towards Robotic Dexterous Hand Intelligence: A Survey","ref_index":155,"is_internal_anchor":true},{"citing_arxiv_id":"2604.09651","citing_title":"FlowHijack: A Dynamics-Aware Backdoor Attack on Flow-Matching Vision-Language-Action Models","ref_index":8,"is_internal_anchor":true},{"citing_arxiv_id":"2605.12090","citing_title":"World Action Models: The Next Frontier in Embodied AI","ref_index":169,"is_internal_anchor":true},{"citing_arxiv_id":"2604.26509","citing_title":"3D Generation for Embodied AI and Robotic Simulation: A Survey","ref_index":190,"is_internal_anchor":true},{"citing_arxiv_id":"2604.26509","citing_title":"3D Generation for Embodied AI and Robotic Simulation: A Survey","ref_index":190,"is_internal_anchor":true},{"citing_arxiv_id":"2604.23001","citing_title":"Vision-Language-Action in Robotics: A Survey of Datasets, Benchmarks, and Data Engines","ref_index":6,"is_internal_anchor":true},{"citing_arxiv_id":"2604.22615","citing_title":"GazeVLA: Learning Human Intention for Robotic Manipulation","ref_index":21,"is_internal_anchor":true},{"citing_arxiv_id":"2604.21241","citing_title":"CorridorVLA: Explicit Spatial Constraints for Generative Action Heads via Sparse Anchors","ref_index":28,"is_internal_anchor":true},{"citing_arxiv_id":"2604.12908","citing_title":"Robotic Manipulation is Vision-to-Geometry Mapping ($f(v) \\rightarrow G$): Vision-Geometry Backbones over Language and Video Models","ref_index":16,"is_internal_anchor":true},{"citing_arxiv_id":"2604.08544","citing_title":"SIM1: Physics-Aligned Simulator as Zero-Shot Data Scaler in Deformable Worlds","ref_index":17,"is_internal_anchor":true},{"citing_arxiv_id":"2506.18088","citing_title":"RoboTwin 2.0: A Scalable Data Generator and Benchmark with Strong Domain Randomization for Robust Bimanual Robotic Manipulation","ref_index":11,"is_internal_anchor":true},{"citing_arxiv_id":"2604.26509","citing_title":"3D Generation for Embodied AI and Robotic Simulation: A Survey","ref_index":190,"is_internal_anchor":true},{"citing_arxiv_id":"2604.13942","citing_title":"Goal2Skill: Long-Horizon Manipulation with Adaptive Planning and Reflection","ref_index":3,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":1,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/6Z5RRGFM332ONHQZVDEEGS642O","json":"https://pith.science/pith/6Z5RRGFM332ONHQZVDEEGS642O.json","graph_json":"https://pith.science/api/pith-number/6Z5RRGFM332ONHQZVDEEGS642O/graph.json","events_json":"https://pith.science/api/pith-number/6Z5RRGFM332ONHQZVDEEGS642O/events.json","paper":"https://pith.science/paper/6Z5RRGFM"},"agent_actions":{"view_html":"https://pith.science/pith/6Z5RRGFM332ONHQZVDEEGS642O","download_json":"https://pith.science/pith/6Z5RRGFM332ONHQZVDEEGS642O.json","view_paper":"https://pith.science/paper/6Z5RRGFM","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2505.03233&json=true","fetch_graph":"https://pith.science/api/pith-number/6Z5RRGFM332ONHQZVDEEGS642O/graph.json","fetch_events":"https://pith.science/api/pith-number/6Z5RRGFM332ONHQZVDEEGS642O/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/6Z5RRGFM332ONHQZVDEEGS642O/action/timestamp_anchor","attest_storage":"https://pith.science/pith/6Z5RRGFM332ONHQZVDEEGS642O/action/storage_attestation","attest_author":"https://pith.science/pith/6Z5RRGFM332ONHQZVDEEGS642O/action/author_attestation","sign_citation":"https://pith.science/pith/6Z5RRGFM332ONHQZVDEEGS642O/action/citation_signature","submit_replication":"https://pith.science/pith/6Z5RRGFM332ONHQZVDEEGS642O/action/replication_record"}},"created_at":"2026-05-17T23:38:13.102696+00:00","updated_at":"2026-05-17T23:38:13.102696+00:00"}