{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:ILI2JTPC2LAVCZ6V2MSBGJ7RUV","short_pith_number":"pith:ILI2JTPC","schema_version":"1.0","canonical_sha256":"42d1a4cde2d2c15167d5d3241327f1a56bc45dbacb36a1d56519d8c5a32a62f1","source":{"kind":"arxiv","id":"2505.18719","version":1},"attestation_state":"computed","paper":{"title":"VLA-RL: Towards Masterful and General Robotic Manipulation with Scalable Reinforcement Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"VLA-RL applies online reinforcement learning to raise pretrained vision-language-action models above finetuned baselines on robot tasks.","cross_cats":["cs.AI"],"primary_cat":"cs.RO","authors_text":"Chubin Zhang, Guanxing Lu, Haonan Jiang, Wenkai Guo, Yansong Tang, Yuheng Zhou, Zifeng Gao, Ziwei Wang","submitted_at":"2025-05-24T14:42:51Z","abstract_excerpt":"Recent high-capacity vision-language-action (VLA) models have demonstrated impressive performance on a range of robotic manipulation tasks by imitating human demonstrations. However, exploiting offline data with limited visited states will cause execution failure in out-of-distribution scenarios. Intuitively, an exploration-based method that improves on online collected data at test time could address this limitation. We present VLA-RL, an algorithmic and systematic framework that leverages online reinforcement learning (RL) to improve pretrained auto-regressive VLAs in downstream tasks. Withi"},"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":false},"canonical_record":{"source":{"id":"2505.18719","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.RO","submitted_at":"2025-05-24T14:42:51Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"b33a2cb43154b37dea226bf00ad80291d2967096372f3629d2434c613d22b641","abstract_canon_sha256":"de19876faf1eddbde8804dc887eb945ac2b1a2b9861aa7505585119aa0f575e2"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:38:47.804302Z","signature_b64":"OG/rfQxiF+S+hkjnZu/PVdI4YdICApn7LWLm+DPUBoTfVh2UwZjfQ0D1N/jJlB4Nzot06MV9OAC3AM59q7uEAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"42d1a4cde2d2c15167d5d3241327f1a56bc45dbacb36a1d56519d8c5a32a62f1","last_reissued_at":"2026-05-17T23:38:47.803799Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:38:47.803799Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"VLA-RL: Towards Masterful and General Robotic Manipulation with Scalable Reinforcement Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"VLA-RL applies online reinforcement learning to raise pretrained vision-language-action models above finetuned baselines on robot tasks.","cross_cats":["cs.AI"],"primary_cat":"cs.RO","authors_text":"Chubin Zhang, Guanxing Lu, Haonan Jiang, Wenkai Guo, Yansong Tang, Yuheng Zhou, Zifeng Gao, Ziwei Wang","submitted_at":"2025-05-24T14:42:51Z","abstract_excerpt":"Recent high-capacity vision-language-action (VLA) models have demonstrated impressive performance on a range of robotic manipulation tasks by imitating human demonstrations. However, exploiting offline data with limited visited states will cause execution failure in out-of-distribution scenarios. Intuitively, an exploration-based method that improves on online collected data at test time could address this limitation. We present VLA-RL, an algorithmic and systematic framework that leverages online reinforcement learning (RL) to improve pretrained auto-regressive VLAs in downstream tasks. Withi"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"VLA-RL enables OpenVLA-7B to surpass the strongest finetuned baseline by 4.5% on 40 challenging robotic manipulation tasks in LIBERO, and even matches the performance of advanced commercial models such as π₀-FAST.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That a vision-language model fine-tuned on pseudo reward labels from automatically extracted task segments will provide sufficiently accurate and generalizable rewards for online RL across diverse out-of-distribution scenarios.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"VLA-RL applies online RL to pretrained VLAs, yielding a 4.5% gain over strong baselines on 40 LIBERO manipulation tasks and matching commercial models like π₀-FAST.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"VLA-RL applies online reinforcement learning to raise pretrained vision-language-action models above finetuned baselines on robot tasks.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"b6ef84180e0065c11bc639260d7b1bd6818e521518951c230cfdd864f83619d7"},"source":{"id":"2505.18719","kind":"arxiv","version":1},"verdict":{"id":"2a57462a-b22a-44d3-90e0-a7abd20e77e6","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-16T12:52:28.368258Z","strongest_claim":"VLA-RL enables OpenVLA-7B to surpass the strongest finetuned baseline by 4.5% on 40 challenging robotic manipulation tasks in LIBERO, and even matches the performance of advanced commercial models such as π₀-FAST.","one_line_summary":"VLA-RL applies online RL to pretrained VLAs, yielding a 4.5% gain over strong baselines on 40 LIBERO manipulation tasks and matching commercial models like π₀-FAST.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That a vision-language model fine-tuned on pseudo reward labels from automatically extracted task segments will provide sufficiently accurate and generalizable rewards for online RL across diverse out-of-distribution scenarios.","pith_extraction_headline":"VLA-RL applies online reinforcement learning to raise pretrained vision-language-action models above finetuned baselines on robot tasks."},"references":{"count":93,"sample":[{"doi":"","year":2022,"title":"Proceedings of Advances in Neural Information Processing Systems (NeurIPS) 35, 28955–28971 (2022) 2, 3","work_id":"2b02057e-69a9-41d7-91b4-09a4ee4adf92","ref_index":1,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":1910,"title":"Solving Rubik's Cube with a Robot Hand","work_id":"81bf9cee-6de8-49f6-b967-3cb853e5ba67","ref_index":2,"cited_arxiv_id":"1910.07113","is_internal_anchor":true},{"doi":"","year":2025,"title":"Univg-r1: Reasoning guided universal visual grounding with reinforcement learning","work_id":"fad4943c-f29d-4daa-a7ba-dd6f03aac5aa","ref_index":3,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2022,"title":"Proceedings of Advances in Neural Information Processing Systems (NeurIPS) 35, 24639– 24654 (2022) 2","work_id":"f271101d-b0ac-4f48-8199-518a4806ce5f","ref_index":4,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2023,"title":"In: Proceedings of International Conference on Machine Learning (ICML)","work_id":"1973ad0d-d7fd-440f-b9b7-60d76e4519de","ref_index":5,"cited_arxiv_id":"","is_internal_anchor":false}],"resolved_work":93,"snapshot_sha256":"e731bc05c403e9713eb1696c70072c04f7e7a544ffe099fa2e2283171345a01a","internal_anchors":36},"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":"2505.18719","created_at":"2026-05-17T23:38:47.803883+00:00"},{"alias_kind":"arxiv_version","alias_value":"2505.18719v1","created_at":"2026-05-17T23:38:47.803883+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.18719","created_at":"2026-05-17T23:38:47.803883+00:00"},{"alias_kind":"pith_short_12","alias_value":"ILI2JTPC2LAV","created_at":"2026-05-18T12:33:37.589309+00:00"},{"alias_kind":"pith_short_16","alias_value":"ILI2JTPC2LAVCZ6V","created_at":"2026-05-18T12:33:37.589309+00:00"},{"alias_kind":"pith_short_8","alias_value":"ILI2JTPC","created_at":"2026-05-18T12:33:37.589309+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":35,"internal_anchor_count":35,"sample":[{"citing_arxiv_id":"2605.22896","citing_title":"Agentic-VLA: Efficient Online Adaptation for Vision-Language-Action Models","ref_index":10,"is_internal_anchor":true},{"citing_arxiv_id":"2510.02590","citing_title":"Use the Online Network If You Can: Towards Fast and Stable Reinforcement Learning","ref_index":4,"is_internal_anchor":true},{"citing_arxiv_id":"2602.10503","citing_title":"Towards Long-Lived Robots: Continual Learning VLA Models via Reinforcement Fine-Tuning","ref_index":44,"is_internal_anchor":true},{"citing_arxiv_id":"2602.09023","citing_title":"TwinRL: Digital Twin-Driven Reinforcement Learning for Real-World Robotic Manipulation","ref_index":38,"is_internal_anchor":true},{"citing_arxiv_id":"2602.18532","citing_title":"VLANeXt: Recipes for Building Strong VLA Models","ref_index":24,"is_internal_anchor":true},{"citing_arxiv_id":"2605.12334","citing_title":"Reinforcing VLAs in Task-Agnostic World Models","ref_index":24,"is_internal_anchor":true},{"citing_arxiv_id":"2605.16154","citing_title":"Learn Where Outcomes Diverge: Efficient VLA RL via Probabilistic Chunk Masking","ref_index":19,"is_internal_anchor":true},{"citing_arxiv_id":"2605.17517","citing_title":"AffordVLA: Injecting Affordance Representations into Vision-Language-Action Models via Implicit Feature Alignment","ref_index":35,"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":129,"is_internal_anchor":true},{"citing_arxiv_id":"2605.19678","citing_title":"RoVLA: Multi-Consistency Constraints for Robust Vision-Language-Action Models","ref_index":37,"is_internal_anchor":true},{"citing_arxiv_id":"2510.12710","citing_title":"Reflection-Based Task Adaptation for Self-Improving VLA","ref_index":23,"is_internal_anchor":true},{"citing_arxiv_id":"2510.17640","citing_title":"RESample: A Robust Data Augmentation Framework via Exploratory Sampling for Robotic Manipulation","ref_index":14,"is_internal_anchor":true},{"citing_arxiv_id":"2511.15669","citing_title":"DeepThinkVLA: Enhancing Reasoning Capability of Vision-Language-Action Models","ref_index":28,"is_internal_anchor":true},{"citing_arxiv_id":"2508.13073","citing_title":"Large VLM-based Vision-Language-Action Models for Robotic Manipulation: A Survey","ref_index":181,"is_internal_anchor":true},{"citing_arxiv_id":"2510.03827","citing_title":"LIBERO-PRO: Towards Robust and Fair Evaluation of Vision-Language-Action Models Beyond Memorization","ref_index":15,"is_internal_anchor":true},{"citing_arxiv_id":"2511.18082","citing_title":"ActDistill: General Action-Guided Self-Derived Distillation for Efficient Vision-Language-Action Models","ref_index":17,"is_internal_anchor":true},{"citing_arxiv_id":"2509.06951","citing_title":"F1: A Vision-Language-Action Model Bridging Understanding and Generation to Actions","ref_index":19,"is_internal_anchor":true},{"citing_arxiv_id":"2602.05765","citing_title":"RL-VLA$^3$: A Flexible and Asynchronous Reinforcement Learning Framework for VLA Training","ref_index":13,"is_internal_anchor":true},{"citing_arxiv_id":"2602.10698","citing_title":"AugVLA-3D: Depth-Driven Feature Augmentation for Vision-Language-Action Models","ref_index":13,"is_internal_anchor":true},{"citing_arxiv_id":"2602.11075","citing_title":"RISE: Self-Improving Robot Policy with Compositional World Model","ref_index":63,"is_internal_anchor":true},{"citing_arxiv_id":"2603.15757","citing_title":"You've Got a Golden Ticket: Improving Generative Robot Policies With A Single Noise Vector","ref_index":18,"is_internal_anchor":true},{"citing_arxiv_id":"2509.09674","citing_title":"SimpleVLA-RL: Scaling VLA Training via Reinforcement Learning","ref_index":43,"is_internal_anchor":true},{"citing_arxiv_id":"2605.11832","citing_title":"Learning Action Manifold with Multi-view Latent Priors for Robotic Manipulation","ref_index":49,"is_internal_anchor":true},{"citing_arxiv_id":"2605.12334","citing_title":"Reinforcing VLAs in Task-Agnostic World Models","ref_index":24,"is_internal_anchor":true},{"citing_arxiv_id":"2511.14759","citing_title":"$\\pi^{*}_{0.6}$: a VLA That Learns From Experience","ref_index":32,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/ILI2JTPC2LAVCZ6V2MSBGJ7RUV","json":"https://pith.science/pith/ILI2JTPC2LAVCZ6V2MSBGJ7RUV.json","graph_json":"https://pith.science/api/pith-number/ILI2JTPC2LAVCZ6V2MSBGJ7RUV/graph.json","events_json":"https://pith.science/api/pith-number/ILI2JTPC2LAVCZ6V2MSBGJ7RUV/events.json","paper":"https://pith.science/paper/ILI2JTPC"},"agent_actions":{"view_html":"https://pith.science/pith/ILI2JTPC2LAVCZ6V2MSBGJ7RUV","download_json":"https://pith.science/pith/ILI2JTPC2LAVCZ6V2MSBGJ7RUV.json","view_paper":"https://pith.science/paper/ILI2JTPC","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2505.18719&json=true","fetch_graph":"https://pith.science/api/pith-number/ILI2JTPC2LAVCZ6V2MSBGJ7RUV/graph.json","fetch_events":"https://pith.science/api/pith-number/ILI2JTPC2LAVCZ6V2MSBGJ7RUV/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ILI2JTPC2LAVCZ6V2MSBGJ7RUV/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ILI2JTPC2LAVCZ6V2MSBGJ7RUV/action/storage_attestation","attest_author":"https://pith.science/pith/ILI2JTPC2LAVCZ6V2MSBGJ7RUV/action/author_attestation","sign_citation":"https://pith.science/pith/ILI2JTPC2LAVCZ6V2MSBGJ7RUV/action/citation_signature","submit_replication":"https://pith.science/pith/ILI2JTPC2LAVCZ6V2MSBGJ7RUV/action/replication_record"}},"created_at":"2026-05-17T23:38:47.803883+00:00","updated_at":"2026-05-17T23:38:47.803883+00:00"}