{"work":{"id":"d8c0c873-b2fc-44a5-a0c8-0d4a698783fb","openalex_id":null,"doi":null,"arxiv_id":"2506.21539","raw_key":null,"title":"WorldVLA: Towards Autoregressive Action World Model","authors":null,"authors_text":"Jun Cen, Chaohui Yu, Hangjie Yuan, Yuming Jiang, Siteng Huang, Jiayan Guo","year":2025,"venue":"cs.RO","abstract":"We present WorldVLA, an autoregressive action world model that unifies action and image understanding and generation. Our WorldVLA intergrates Vision-Language-Action (VLA) model and world model in one single framework. The world model predicts future images by leveraging both action and image understanding, with the purpose of learning the underlying physics of the environment to improve action generation. Meanwhile, the action model generates the subsequent actions based on image observations, aiding in visual understanding and in turn helps visual generation of the world model. We demonstrate that WorldVLA outperforms standalone action and world models, highlighting the mutual enhancement between the world model and the action model. In addition, we find that the performance of the action model deteriorates when generating sequences of actions in an autoregressive manner. This phenomenon can be attributed to the model's limited generalization capability for action prediction, leading to the propagation of errors from earlier actions to subsequent ones. To address this issue, we propose an attention mask strategy that selectively masks prior actions during the generation of the current action, which shows significant performance improvement in the action chunk generation task.","external_url":"https://arxiv.org/abs/2506.21539","cited_by_count":null,"metadata_source":"pith","metadata_fetched_at":"2026-05-24T01:25:54.361613+00:00","pith_arxiv_id":"2506.21539","created_at":"2026-05-09T06:55:43.930119+00:00","updated_at":"2026-06-05T21:23:00.469572+00:00","title_quality_ok":true,"display_title":"WorldVLA: Towards Autoregressive Action World Model","render_title":"WorldVLA: Towards Autoregressive Action World Model"},"hub":{"state":{"work_id":"d8c0c873-b2fc-44a5-a0c8-0d4a698783fb","tier":"hub","tier_reason":"10+ Pith inbound or 1,000+ external citations","pith_inbound_count":62,"external_cited_by_count":null,"distinct_field_count":6,"first_pith_cited_at":"2024-05-23T01:43:54+00:00","last_pith_cited_at":"2026-05-21T13:13:31+00:00","author_build_status":"not_needed","summary_status":"needed","contexts_status":"needed","graph_status":"needed","ask_index_status":"not_needed","reader_status":"not_needed","recognition_status":"not_needed","updated_at":"2026-06-06T08:10:29.710048+00:00","tier_text":"hub"},"tier":"hub","role_counts":[{"context_role":"background","n":25},{"context_role":"baseline","n":4},{"context_role":"method","n":1}],"polarity_counts":[{"context_polarity":"background","n":25},{"context_polarity":"baseline","n":4},{"context_polarity":"use_method","n":1}],"runs":{"context_extract":{"job_type":"context_extract","status":"succeeded","result":{"enqueued_papers":25},"error":null,"updated_at":"2026-05-14T17:48:39.541987+00:00"},"graph_features":{"job_type":"graph_features","status":"succeeded","result":{"co_cited":[{"title":"$\\pi_0$: A Vision-Language-Action Flow Model for General Robot Control","work_id":"f790abdc-a796-482f-a40d-f8ee035ecfc2","shared_citers":22},{"title":"Fine-Tuning Vision-Language-Action Models: Optimizing Speed and Success","work_id":"04f46bb3-4346-47e8-bf09-c75d91f96e87","shared_citers":20},{"title":"OpenVLA: An Open-Source Vision-Language-Action Model","work_id":"3e7e65c5-5aed-4fe9-8414-2092bcb31cc7","shared_citers":19},{"title":"GR00T N1: An Open Foundation Model for Generalist Humanoid Robots","work_id":"e2db69c7-ee8a-4cb7-a761-7b8de1dfcf97","shared_citers":17},{"title":"$\\pi_{0.5}$: a Vision-Language-Action Model with Open-World Generalization","work_id":"d1ad7304-d09a-49bc-809e-846439f6aff9","shared_citers":16},{"title":"FAST: Efficient Action Tokenization for Vision-Language-Action Models","work_id":"83a8f966-6cfa-4f21-81f3-87440aae238f","shared_citers":15},{"title":"RT-1: Robotics Transformer for Real-World Control at Scale","work_id":"e11bda85-8531-46bc-a07f-d0ade3643ab1","shared_citers":13},{"title":"LIBERO-Plus: In-depth Robustness Analysis of Vision-Language-Action Models","work_id":"e35c8c6d-977d-4af1-963a-766ba98703ce","shared_citers":12},{"title":"Wan: Open and Advanced Large-Scale Video Generative Models","work_id":"ad3ebc3b-4224-46c9-b61d-bcf135da0a7c","shared_citers":12},{"title":"SpatialVLA: Exploring Spatial Representations for Visual-Language-Action Model","work_id":"592041b3-3ca2-4836-8dd4-f8095d8a692b","shared_citers":11},{"title":"UniVLA: Learning to Act Anywhere with Task-centric Latent Actions","work_id":"e05d654d-db73-48f6-9318-381b6798bac9","shared_citers":11},{"title":"Learning Fine-Grained Bimanual Manipulation with Low-Cost Hardware","work_id":"6fe159e0-fa73-481a-88d4-4719c15140be","shared_citers":10},{"title":"Octo: An Open-Source Generalist Robot Policy","work_id":"f9ca0722-8855-48c3-a27a-0eefb7e19253","shared_citers":10},{"title":"Causal World Modeling for Robot Control","work_id":"a33c4ee0-db06-4f9a-8852-c62e3a72fc27","shared_citers":9},{"title":"CogACT: A Foundational Vision-Language-Action Model for Synergizing Cognition and Action in Robotic Manipulation","work_id":"4b158d3e-3dff-4412-85cd-baa879465a5e","shared_citers":9},{"title":"Motus: A Unified Latent Action World Model","work_id":"d0b2d257-524d-4d67-9daf-5fb43e5e977a","shared_citers":9},{"title":"SmolVLA: A Vision-Language-Action Model for Affordable and Efficient Robotics","work_id":"0c5e9314-5fa7-4613-ad12-605a71d561d2","shared_citers":9},{"title":"X-VLA: Soft-Prompted Transformer as Scalable Cross-Embodiment Vision-Language-Action Model","work_id":"13faca8d-e96d-4e6c-a441-9f2683d11934","shared_citers":9},{"title":"Cosmos Policy: Fine-Tuning Video Models for Visuomotor Control and Planning","work_id":"3d63039f-41b0-4a31-af31-6fc10f5c1b1b","shared_citers":8},{"title":"Dreamvla: A vision- language-action model dreamed with comprehensive world knowledge.CoRR, abs/2507.04447","work_id":"830b1108-8122-404f-a791-f926fbc48669","shared_citers":8},{"title":"Fast-WAM: Do World Action Models Need Test-time Future Imagination?","work_id":"772d0226-9ad6-42c0-9c79-d36d37edbbe4","shared_citers":8},{"title":"Qwen3-VL Technical Report","work_id":"1fe243aa-e3c0-4da6-b391-4cbcfc88d5c0","shared_citers":8},{"title":"RoboTwin 2.0: A Scalable Data Generator and Benchmark with Strong Domain Randomization for Robust Bimanual Robotic Manipulation","work_id":"9b985126-4a2f-4bdf-b014-2a7524ec634e","shared_citers":8},{"title":"World Action Models are Zero-shot Policies","work_id":"9a85fc69-74df-450e-94cd-69d186e9e830","shared_citers":8}],"time_series":[{"n":38,"year":2026}],"dependency_candidates":[]},"error":null,"updated_at":"2026-05-14T17:49:14.292152+00:00"},"identity_refresh":{"job_type":"identity_refresh","status":"succeeded","result":{"items":[{"title":"Qwen3 Technical Report","outcome":"unchanged","work_id":"25a4e30c-1232-48e7-9925-02fa12ba7c9e","resolver":"local_arxiv","confidence":0.98,"old_work_id":"25a4e30c-1232-48e7-9925-02fa12ba7c9e"}],"counts":{"fixed":0,"merged":0,"unchanged":1,"quarantined":0,"needs_external_resolution":0},"errors":[],"attempted":1},"error":null,"updated_at":"2026-05-14T17:48:36.370980+00:00"},"summary_claims":{"job_type":"summary_claims","status":"succeeded","result":{"title":"WorldVLA: Towards Autoregressive Action World Model","claims":[{"claim_text":"We present WorldVLA, an autoregressive action world model that unifies action and image understanding and generation. Our WorldVLA intergrates Vision-Language-Action (VLA) model and world model in one single framework. The world model predicts future images by leveraging both action and image understanding, with the purpose of learning the underlying physics of the environment to improve action generation. Meanwhile, the action model generates the subsequent actions based on image observations, aiding in visual understanding and in turn helps visual generation of the world model. We demonstrat","claim_type":"abstract","evidence_strength":"source_metadata"}],"why_cited":"Pith tracks WorldVLA: Towards Autoregressive Action World Model because it crossed a citation-hub threshold.","role_counts":[]},"error":null,"updated_at":"2026-05-14T17:48:58.554135+00:00"}},"summary":{"title":"WorldVLA: Towards Autoregressive Action World Model","claims":[{"claim_text":"We present WorldVLA, an autoregressive action world model that unifies action and image understanding and generation. Our WorldVLA intergrates Vision-Language-Action (VLA) model and world model in one single framework. The world model predicts future images by leveraging both action and image understanding, with the purpose of learning the underlying physics of the environment to improve action generation. Meanwhile, the action model generates the subsequent actions based on image observations, aiding in visual understanding and in turn helps visual generation of the world model. We demonstrat","claim_type":"abstract","evidence_strength":"source_metadata"}],"why_cited":"Pith tracks WorldVLA: Towards Autoregressive Action World Model because it crossed a citation-hub threshold.","role_counts":[]},"graph":{"co_cited":[{"title":"$\\pi_0$: A Vision-Language-Action Flow Model for General Robot Control","work_id":"f790abdc-a796-482f-a40d-f8ee035ecfc2","shared_citers":22},{"title":"Fine-Tuning Vision-Language-Action Models: Optimizing Speed and Success","work_id":"04f46bb3-4346-47e8-bf09-c75d91f96e87","shared_citers":20},{"title":"OpenVLA: An Open-Source Vision-Language-Action Model","work_id":"3e7e65c5-5aed-4fe9-8414-2092bcb31cc7","shared_citers":19},{"title":"GR00T N1: An Open Foundation Model for Generalist Humanoid Robots","work_id":"e2db69c7-ee8a-4cb7-a761-7b8de1dfcf97","shared_citers":17},{"title":"$\\pi_{0.5}$: a Vision-Language-Action Model with Open-World Generalization","work_id":"d1ad7304-d09a-49bc-809e-846439f6aff9","shared_citers":16},{"title":"FAST: Efficient Action Tokenization for Vision-Language-Action Models","work_id":"83a8f966-6cfa-4f21-81f3-87440aae238f","shared_citers":15},{"title":"RT-1: Robotics Transformer for Real-World Control at Scale","work_id":"e11bda85-8531-46bc-a07f-d0ade3643ab1","shared_citers":13},{"title":"LIBERO-Plus: In-depth Robustness Analysis of Vision-Language-Action Models","work_id":"e35c8c6d-977d-4af1-963a-766ba98703ce","shared_citers":12},{"title":"Wan: Open and Advanced Large-Scale Video Generative Models","work_id":"ad3ebc3b-4224-46c9-b61d-bcf135da0a7c","shared_citers":12},{"title":"SpatialVLA: Exploring Spatial Representations for Visual-Language-Action Model","work_id":"592041b3-3ca2-4836-8dd4-f8095d8a692b","shared_citers":11},{"title":"UniVLA: Learning to Act Anywhere with Task-centric Latent Actions","work_id":"e05d654d-db73-48f6-9318-381b6798bac9","shared_citers":11},{"title":"Learning Fine-Grained Bimanual Manipulation with Low-Cost Hardware","work_id":"6fe159e0-fa73-481a-88d4-4719c15140be","shared_citers":10},{"title":"Octo: An Open-Source Generalist Robot Policy","work_id":"f9ca0722-8855-48c3-a27a-0eefb7e19253","shared_citers":10},{"title":"Causal World Modeling for Robot Control","work_id":"a33c4ee0-db06-4f9a-8852-c62e3a72fc27","shared_citers":9},{"title":"CogACT: A Foundational Vision-Language-Action Model for Synergizing Cognition and Action in Robotic Manipulation","work_id":"4b158d3e-3dff-4412-85cd-baa879465a5e","shared_citers":9},{"title":"Motus: A Unified Latent Action World Model","work_id":"d0b2d257-524d-4d67-9daf-5fb43e5e977a","shared_citers":9},{"title":"SmolVLA: A Vision-Language-Action Model for Affordable and Efficient Robotics","work_id":"0c5e9314-5fa7-4613-ad12-605a71d561d2","shared_citers":9},{"title":"X-VLA: Soft-Prompted Transformer as Scalable Cross-Embodiment Vision-Language-Action Model","work_id":"13faca8d-e96d-4e6c-a441-9f2683d11934","shared_citers":9},{"title":"Cosmos Policy: Fine-Tuning Video Models for Visuomotor Control and Planning","work_id":"3d63039f-41b0-4a31-af31-6fc10f5c1b1b","shared_citers":8},{"title":"Dreamvla: A vision- language-action model dreamed with comprehensive world knowledge.CoRR, abs/2507.04447","work_id":"830b1108-8122-404f-a791-f926fbc48669","shared_citers":8},{"title":"Fast-WAM: Do World Action Models Need Test-time Future Imagination?","work_id":"772d0226-9ad6-42c0-9c79-d36d37edbbe4","shared_citers":8},{"title":"Qwen3-VL Technical Report","work_id":"1fe243aa-e3c0-4da6-b391-4cbcfc88d5c0","shared_citers":8},{"title":"RoboTwin 2.0: A Scalable Data Generator and Benchmark with Strong Domain Randomization for Robust Bimanual Robotic Manipulation","work_id":"9b985126-4a2f-4bdf-b014-2a7524ec634e","shared_citers":8},{"title":"World Action Models are Zero-shot Policies","work_id":"9a85fc69-74df-450e-94cd-69d186e9e830","shared_citers":8}],"time_series":[{"n":38,"year":2026}],"dependency_candidates":[]},"authors":[]}}