{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:E3T33REDAUK72EVJZT4BVCFQGB","short_pith_number":"pith:E3T33RED","schema_version":"1.0","canonical_sha256":"26e7bdc4830515fd12a9ccf81a88b0304845d02a7ee8022d7163fbd4f8dbd609","source":{"kind":"arxiv","id":"2412.14058","version":4},"attestation_state":"computed","paper":{"title":"What Matters in Building Vision-Language-Action Models for Generalist Robots","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"Specific choices in backbones, architectures, and data timing let simple Vision-Language-Action models set new robot manipulation records.","cross_cats":["cs.CV"],"primary_cat":"cs.RO","authors_text":"Bingyi Kang, Di Guo, Dong Wang, Hanbo Zhang, Huaping Liu, Jirong Liu, Long Qian, Minghuan Liu, Peiyan Li, Tao Kong, Xiao Ma, Xinghang Li, Xinlong Wang","submitted_at":"2024-12-18T17:07:20Z","abstract_excerpt":"To utilize Foundation Vision Language Models (VLMs) for robotic tasks and motion planning, the community has proposed different methods for injecting action components into VLMs and building the Vision-Language-Action models (VLAs). In this work, we disclose the key factors that significantly influence the performance of VLA on robot manipulation problems and focus on answering three essential design choices: which backbone to select, how to formulate the VLA architectures, and when to add cross-embodiment data. The obtained results convince us firmly to explain why we prefer VLA and develop a"},"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":"2412.14058","kind":"arxiv","version":4},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.RO","submitted_at":"2024-12-18T17:07:20Z","cross_cats_sorted":["cs.CV"],"title_canon_sha256":"68aff8c3e3373b6aae6aedffb5e6c987e6510dab548baddf2b11e15244a28776","abstract_canon_sha256":"334f8ebbf897361337c756e401d8bad05d9afe3016185d1262c03aca70045cd6"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:38:13.007578Z","signature_b64":"u7E0Ynleg5WCOorRf+OVtExejtEqvST/Q+IRlKCL7t+Ru3aj9mn15qdVAl80zfURZ5r4e645ahwLEMHQQfk/AQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"26e7bdc4830515fd12a9ccf81a88b0304845d02a7ee8022d7163fbd4f8dbd609","last_reissued_at":"2026-05-17T23:38:13.006948Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:38:13.006948Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"What Matters in Building Vision-Language-Action Models for Generalist Robots","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"Specific choices in backbones, architectures, and data timing let simple Vision-Language-Action models set new robot manipulation records.","cross_cats":["cs.CV"],"primary_cat":"cs.RO","authors_text":"Bingyi Kang, Di Guo, Dong Wang, Hanbo Zhang, Huaping Liu, Jirong Liu, Long Qian, Minghuan Liu, Peiyan Li, Tao Kong, Xiao Ma, Xinghang Li, Xinlong Wang","submitted_at":"2024-12-18T17:07:20Z","abstract_excerpt":"To utilize Foundation Vision Language Models (VLMs) for robotic tasks and motion planning, the community has proposed different methods for injecting action components into VLMs and building the Vision-Language-Action models (VLAs). In this work, we disclose the key factors that significantly influence the performance of VLA on robot manipulation problems and focus on answering three essential design choices: which backbone to select, how to formulate the VLA architectures, and when to add cross-embodiment data. The obtained results convince us firmly to explain why we prefer VLA and develop a"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"Through extensive experiments which include over 8 VLM backbones, 4 policy architectures, and over 600 distinct designed experiments, we provide a detailed guidebook for the future design of VLAs... RoboVLMs... achieve a new state-of-the-art performance in three simulation tasks and real-world experiments.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That the chosen simulation tasks and real-world experiments are representative enough of generalist robot manipulation that the observed ranking of design choices will transfer to new tasks, embodiments, and environments not tested in the 600+ runs.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"Systematic tests of VLM backbones, policy architectures, and cross-embodiment data yield RoboVLMs that set new SOTA on robot manipulation benchmarks while requiring few manual designs.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Specific choices in backbones, architectures, and data timing let simple Vision-Language-Action models set new robot manipulation records.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"233bf637abbb829875db7cf18e555ed55f389674e64b32d2597d2ef338c8487f"},"source":{"id":"2412.14058","kind":"arxiv","version":4},"verdict":{"id":"5de490e9-804b-4794-af7a-10c5bab1f587","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-17T21:33:48.446845Z","strongest_claim":"Through extensive experiments which include over 8 VLM backbones, 4 policy architectures, and over 600 distinct designed experiments, we provide a detailed guidebook for the future design of VLAs... RoboVLMs... achieve a new state-of-the-art performance in three simulation tasks and real-world experiments.","one_line_summary":"Systematic tests of VLM backbones, policy architectures, and cross-embodiment data yield RoboVLMs that set new SOTA on robot manipulation benchmarks while requiring few manual designs.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That the chosen simulation tasks and real-world experiments are representative enough of generalist robot manipulation that the observed ranking of design choices will transfer to new tasks, embodiments, and environments not tested in the 600+ runs.","pith_extraction_headline":"Specific choices in backbones, architectures, and data timing let simple Vision-Language-Action models set new robot manipulation records."},"references":{"count":54,"sample":[{"doi":"","year":2022,"title":"Flamingo: a visual language model for few-shot learning","work_id":"abe0dc9d-38b0-4f78-87e4-1f56476c29c8","ref_index":1,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2023,"title":"Qwen-VL: A Versatile Vision-Language Model for Understanding, Localization, Text Reading, and Beyond","work_id":"cbc2bb21-b6bb-46c0-80bf-107e195ffe10","ref_index":2,"cited_arxiv_id":"2308.12966","is_internal_anchor":true},{"doi":"","year":2024,"title":"PaliGemma: A versatile 3B VLM for transfer","work_id":"df6f48b3-5792-47c7-9614-cb856ea31ad9","ref_index":3,"cited_arxiv_id":"2407.07726","is_internal_anchor":true},{"doi":"","year":2024,"title":"$\\pi_0$: A Vision-Language-Action Flow Model for General Robot Control","work_id":"f790abdc-a796-482f-a40d-f8ee035ecfc2","ref_index":4,"cited_arxiv_id":"2410.24164","is_internal_anchor":true},{"doi":"","year":2023,"title":"RoboCat : A self-improving foundation agent for robotic manipulation","work_id":"143e7731-0488-4088-8e23-63f9d4140118","ref_index":5,"cited_arxiv_id":"","is_internal_anchor":false}],"resolved_work":54,"snapshot_sha256":"068783adc0418b56d569a8d0f13ce528ebe66b651ad9e79116787bc514b71dec","internal_anchors":25},"formal_canon":{"evidence_count":2,"snapshot_sha256":"8df15057d85122e96a54c885c04271576961ff57df3c998dfacce45ccdae174c"},"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":"2412.14058","created_at":"2026-05-17T23:38:13.007049+00:00"},{"alias_kind":"arxiv_version","alias_value":"2412.14058v4","created_at":"2026-05-17T23:38:13.007049+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2412.14058","created_at":"2026-05-17T23:38:13.007049+00:00"},{"alias_kind":"pith_short_12","alias_value":"E3T33REDAUK7","created_at":"2026-05-18T12:33:37.589309+00:00"},{"alias_kind":"pith_short_16","alias_value":"E3T33REDAUK72EVJ","created_at":"2026-05-18T12:33:37.589309+00:00"},{"alias_kind":"pith_short_8","alias_value":"E3T33RED","created_at":"2026-05-18T12:33:37.589309+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":20,"internal_anchor_count":20,"sample":[{"citing_arxiv_id":"2505.17015","citing_title":"Multi-SpatialMLLM: Multi-Frame Spatial Understanding with Multi-Modal Large Language Models","ref_index":37,"is_internal_anchor":true},{"citing_arxiv_id":"2605.15298","citing_title":"PhysBrain 1.0 Technical Report","ref_index":22,"is_internal_anchor":true},{"citing_arxiv_id":"2511.14148","citing_title":"AsyncVLA: Asynchronous Flow Matching for Vision-Language-Action Models","ref_index":39,"is_internal_anchor":true},{"citing_arxiv_id":"2507.15493","citing_title":"GR-3 Technical Report","ref_index":44,"is_internal_anchor":true},{"citing_arxiv_id":"2512.09928","citing_title":"HiF-VLA: Hindsight, Insight and Foresight through Motion Representation for Vision-Language-Action Models","ref_index":26,"is_internal_anchor":true},{"citing_arxiv_id":"2507.04447","citing_title":"DreamVLA: A Vision-Language-Action Model Dreamed with Comprehensive World Knowledge","ref_index":37,"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":4,"is_internal_anchor":true},{"citing_arxiv_id":"2507.23682","citing_title":"villa-X: Enhancing Latent Action Modeling in Vision-Language-Action Models","ref_index":34,"is_internal_anchor":true},{"citing_arxiv_id":"2602.20323","citing_title":"PhysMem: Scaling Test-Time Memory for Embodied Physical Reasoning","ref_index":37,"is_internal_anchor":true},{"citing_arxiv_id":"2605.13403","citing_title":"RotVLA: Rotational Latent Action for Vision-Language-Action Model","ref_index":14,"is_internal_anchor":true},{"citing_arxiv_id":"2604.03181","citing_title":"Multi-View Video Diffusion Policy: A 3D Spatio-Temporal-Aware Video Action Model","ref_index":21,"is_internal_anchor":true},{"citing_arxiv_id":"2605.12167","citing_title":"From Imagined Futures to Executable Actions: Mixture of Latent Actions for Robot Manipulation","ref_index":30,"is_internal_anchor":true},{"citing_arxiv_id":"2505.06111","citing_title":"UniVLA: Learning to Act Anywhere with Task-centric Latent Actions","ref_index":45,"is_internal_anchor":true},{"citing_arxiv_id":"2605.09537","citing_title":"Drift is a Sampling Error: SNR-Aware Power Distributions for Long-Horizon Robotic Planning","ref_index":1,"is_internal_anchor":true},{"citing_arxiv_id":"2501.15830","citing_title":"SpatialVLA: Exploring Spatial Representations for Visual-Language-Action Model","ref_index":33,"is_internal_anchor":true},{"citing_arxiv_id":"2605.06175","citing_title":"VLA-GSE: Boosting Parameter-Efficient Fine-Tuning in VLA with Generalized and Specialized Experts","ref_index":23,"is_internal_anchor":true},{"citing_arxiv_id":"2605.06175","citing_title":"VLA-GSE: Boosting Parameter-Efficient Fine-Tuning in VLA with Generalized and Specialized Experts","ref_index":19,"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":21,"is_internal_anchor":true},{"citing_arxiv_id":"2604.20100","citing_title":"JoyAI-RA 0.1: A Foundation Model for Robotic Autonomy","ref_index":25,"is_internal_anchor":true},{"citing_arxiv_id":"2604.20348","citing_title":"Bimanual Robot Manipulation via Multi-Agent In-Context Learning","ref_index":27,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":2,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/E3T33REDAUK72EVJZT4BVCFQGB","json":"https://pith.science/pith/E3T33REDAUK72EVJZT4BVCFQGB.json","graph_json":"https://pith.science/api/pith-number/E3T33REDAUK72EVJZT4BVCFQGB/graph.json","events_json":"https://pith.science/api/pith-number/E3T33REDAUK72EVJZT4BVCFQGB/events.json","paper":"https://pith.science/paper/E3T33RED"},"agent_actions":{"view_html":"https://pith.science/pith/E3T33REDAUK72EVJZT4BVCFQGB","download_json":"https://pith.science/pith/E3T33REDAUK72EVJZT4BVCFQGB.json","view_paper":"https://pith.science/paper/E3T33RED","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2412.14058&json=true","fetch_graph":"https://pith.science/api/pith-number/E3T33REDAUK72EVJZT4BVCFQGB/graph.json","fetch_events":"https://pith.science/api/pith-number/E3T33REDAUK72EVJZT4BVCFQGB/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/E3T33REDAUK72EVJZT4BVCFQGB/action/timestamp_anchor","attest_storage":"https://pith.science/pith/E3T33REDAUK72EVJZT4BVCFQGB/action/storage_attestation","attest_author":"https://pith.science/pith/E3T33REDAUK72EVJZT4BVCFQGB/action/author_attestation","sign_citation":"https://pith.science/pith/E3T33REDAUK72EVJZT4BVCFQGB/action/citation_signature","submit_replication":"https://pith.science/pith/E3T33REDAUK72EVJZT4BVCFQGB/action/replication_record"}},"created_at":"2026-05-17T23:38:13.007049+00:00","updated_at":"2026-05-17T23:38:13.007049+00:00"}