mlsanityData to 7 Oct 2026

X-VLA: Soft-Prompted Transformer as Scalable Cross-Embodiment Vision-Language-Action Model

arXiv 2510.10274 ↗First posted 11 Oct 2025cs.RO cs.AI cs.CVcited by 285

On the map: Vision-language-action models → Generalist VLA models and benchmarks

Abstract (arXiv)

Successful generalist Vision-Language-Action (VLA) models rely on effective training across diverse robotic platforms with large-scale, cross-embodiment, heterogeneous datasets. To facilitate and leverage the heterogeneity in rich, diverse robotic data sources, we propose a novel Soft Prompt approach with minimally added parameters, by infusing prompt learning concepts into cross-embodiment robot learning and introducing separate sets of learnable embeddings for each distinct data source. These embeddings serve as embodiment-specific prompts, which in unity empower VLA models with effective exploitation of varying cross-embodiment features. Our new X-VLA, a neat flow-matching-based VLA architecture, relies exclusively on soft-prompted standard Transformer encoders, enjoying both scalability and simplicity. Evaluated across 6 simulations as well as 3 real-world robots, our 0.9B instantiation-X-VLA-0.9B simultaneously achieves SOTA performance over a sweep of benchmarks, demonstrating superior results on a wide axes of capabilities, from flexible dexterity to quick adaptation across embodiments, environments, and tasks. Website: https://thu-air-dream.github.io/X-VLA/

Key facts. Cited by 285 papers in the map (data through 7 Oct 2026). Placed in Generalist VLA models and benchmarks, one of 13 directions in Vision-language-action models. Builds on 82 of its 90 references that are in the map. First posted on arXiv 11 Oct 2025.

Builds on

82 of 90 references.

Closest ancestors · 10

Same direction, newest first

  1. 2025PhysiAgent: An Embodied Agent Framework in Physical World29 Sep
  2. 2025FLOWER: Democratizing Generalist Robot Policies with Efficient Vision-Language-Action Flow Policies5 Sep
  3. 2025Unified Vision-Language-Action Model24 Jun
  4. 2025SmolVLA: A Vision-Language-Action Model for Affordable and Efficient Robotics2 Jun
  5. 2025Knowledge Insulating Vision-Language-Action Models: Train Fast, Run Fast, Generalize Better29 May
  6. 2025π0.5: a Vision-Language-Action Model with Open-World Generalization22 Apr
  7. 2025Fine-Tuning Vision-Language-Action Models: Optimizing Speed and Success27 Feb
  8. 2024π0: A Vision-Language-Action Flow Model for General Robot Control31 Oct
  9. 2024OpenVLA: An Open-Source Vision-Language-Action Model13 Jun
  10. 2023RT-2: Vision-Language-Action Models Transfer Web Knowledge to Robotic Control28 Jul

Other references · 72

  1. 2023GPT-4 Technical Reportcited by 18,132
  2. 2023Visual Instruction Tuningcited by 7,792
  3. 2022Scalable Diffusion Models with Transformerscited by 5,931
  4. 2022Flow Matching for Generative Modelingcited by 5,426
  5. 2024Scaling Rectified Flow Transformers for High-Resolution Image Synthesiscited by 4,094
  6. 2021Prefix-Tuning: Optimizing Continuous Prompts for Generationcited by 3,878
  7. 2021The Power of Scale for Parameter-Efficient Prompt Tuningcited by 3,769
  8. 2017Outrageously Large Neural Networks: The Sparsely-Gated Mixture-of-Experts Layercited by 3,413
  9. 2023Qwen Technical Reportcited by 3,273
  10. 2023Diffusion Policy: Visuomotor Policy Learning via Action Diffusioncited by 2,730
  11. 2024LLaVA-OneVision: Easy Visual Task Transfercited by 2,669
  12. 2021Pre-train, Prompt, and Predict: A Systematic Survey of Prompting Methods in Natural Language Processingcited by 2,222
38 more
  1. 2022RT-1: Robotics Transformer for Real-World Control at Scalecited by 2,007
  2. 2023Learning Fine-Grained Bimanual Manipulation with Low-Cost Hardwarecited by 1,490
  3. 2021MLP-Mixer: An all-MLP Architecture for Visioncited by 1,441
  4. 2024Octo: An Open-Source Generalist Robot Policycited by 1,427
  5. 2023LIBERO: Benchmarking Knowledge Transfer for Lifelong Robot Learningcited by 1,338
  6. 2018On the Continuity of Rotation Representations in Neural Networkscited by 1,186
  7. 2025GR00T N1: An Open Foundation Model for Generalist Humanoid Robotscited by 1,167
  8. 2022Planning-oriented Autonomous Drivingcited by 1,068
  9. 2021Learning to Prompt for Continual Learningcited by 938
  10. 2024Gemma: Open Models Based on Gemini Research and Technologycited by 889
  11. 2024DROID: A Large-Scale In-The-Wild Robot Manipulation Datasetcited by 801
  12. 2022MaPLe: Multi-modal Prompt Learningcited by 685
  13. 2024RDT-1B: a Diffusion Foundation Model for Bimanual Manipulationcited by 670
  14. 2025FAST: Efficient Action Tokenization for Vision-Language-Action Modelscited by 643
  15. 2023BridgeData V2: A Dataset for Robot Learning at Scalecited by 634
  16. 2021CALVIN: A Benchmark for Language-Conditioned Policy Learning for Long-Horizon Robot Manipulation Taskscited by 564
  17. 2024Evaluating Real-World Robot Manipulation Policies in Simulationcited by 484
  18. 2022TransFuser: Imitation with Transformer-Based Sensor Fusion for Autonomous Drivingcited by 477
  19. 2025RoboTwin 2.0: A Scalable Data Generator and Benchmark with Strong Domain Randomization for Robust Bimanual Robotic Manipulationcited by 477
  20. 2023Florence-2: Advancing a Unified Representation for a Variety of Vision Taskscited by 403
  21. 2025AgiBot World Colosseo: A Large-scale Manipulation Platform for Scalable and Intelligent Embodied Systemscited by 390
  22. 2023Unleashing Large-Scale Video Generative Pre-training for Visual Robot Manipulationcited by 353
  23. 2024NAVSIM: Data-Driven Non-Reactive Autonomous Vehicle Simulation and Benchmarkingcited by 343
  24. 2024Robotic Control via Embodied Chain-of-Thought Reasoningcited by 340
  25. 20243D-VLA: A 3D Vision-Language-Action Generative World Modelcited by 320
  26. 2024Latent Action Pretraining from Videoscited by 302
  27. 2024TraceVLA: Visual Trace Prompting Enhances Spatial-Temporal Awareness for Generalist Robotic Policiescited by 298
  28. 2024Flow Matching Guide and Codecited by 291
  29. 2023EmbodiedGPT: Vision-Language Pre-Training via Embodied Chain of Thoughtcited by 276
  30. 2023Zero-Shot Robotic Manipulation with Pretrained Image-Editing Diffusion Modelscited by 264
  31. 2017One-Shot Visual Imitation Learning via Meta-Learningcited by 225
  32. 2024Auxiliary-Loss-Free Load Balancing Strategy for Mixture-of-Expertscited by 191
  33. 2025MemoryVLA: Perceptual-Cognitive Memory in Vision-Language-Action Models for Robotic Manipulationcited by 183
  34. 2024VLABench: A Large-Scale Benchmark for Language-Conditioned Robotics Manipulation with Long-Horizon Reasoning Taskscited by 171
  35. 2024RoboMIND: Benchmark on Multi-embodiment Intelligence Normative Data for Robot Manipulationcited by 166
  36. 2024Predictive Inverse Dynamics Models are Scalable Learners for Robotic Manipulationcited by 159
  37. 2025ThinkAct: Vision-Language-Action Reasoning via Reinforced Visual Latent Planningcited by 152
  38. 2024Scaling Proprioceptive-Visual Learning with Heterogeneous Pre-trained Transformerscited by 145

50 most cited of 72 All 72 (JSON)

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 13; 2026: 272.

285 papers.

Sort:

By area:

  1. 2026Causal World Modeling for Robot Controlcited by 289
  2. 2025Motus: A Unified Latent Action World Modelcited by 231
  3. 2026GigaWorld-Policy: An Efficient Action-Centered World--Action Modelcited by 129
  4. 2026π0.7: a Steerable Generalist Robotic Foundation Model with Emergent Capabilitiescited by 127
  5. 2026StarVLA: A Lego-like Codebase for Vision-Language-Action Model Developingcited by 103
  6. 2026ABot-M0: VLA Foundation Model for Robotic Manipulation with Action Manifold Learningcited by 94
  7. 2025RynnVLA-002: A Unified Vision-Language-Action and World Modelcited by 70
  8. 2026Being-H0.7: A Latent World-Action Model from Egocentric Videoscited by 68
  9. 2026Being-H0.5: Scaling Human-Centric Robot Learning for Cross-Embodiment Generalizationcited by 61
  10. 2026Motubrain: An Advanced World Action Model for Robot Controlcited by 54
  11. 2026InternVLA-A1: Unifying Understanding, Generation and Action for Robotic Manipulationcited by 48
  12. 2026Qwen-RobotManip Technical Report: Alignment Unlocks Scale for Robotic Manipulation Foundation Modelscited by 47
38 more
  1. 2025InternData-A1: Pioneering High-Fidelity Synthetic Data for Pre-training Generalist Policycited by 46
  2. 2026WoVR: World Models as Reliable Simulators for Post-Training VLA Policies with RLcited by 46
  3. 2026MolmoAct2: Action Reasoning Models for Real-world Deploymentcited by 45
  4. 2026Qwen-VLA: Unifying Vision-Language-Action Modeling across Tasks, Environments, and Robot Embodimentscited by 42
  5. 2026AHA-WAM:Asynchronous Horizon-Adaptive World-Action Modeling with Observation-Guided Context Routingcited by 41
  6. 2026RMBench: Memory-Dependent Robotic Manipulation Benchmark with Insights into Policy Designcited by 40
  7. 2025Robo-Dopamine: General Process Reward Modeling for High-Precision Robotic Manipulationcited by 37
  8. 2026Harness VLA: Steering Frozen VLAs into Reliable Manipulation Primitives via Memory-Guided Agentscited by 37
  9. 2025Emergence of Human to Robot Transfer in Vision-Language-Action Modelscited by 36
  10. 2026Do World Action Models Generalize Better than VLAs? A Robustness Studycited by 35
  11. 2026RoboDojo: A Unified Sim-and-Real Benchmark for Comprehensive Evaluation of Generalist Robot Manipulation Policiescited by 35
  12. 2026Native Video-Action Pretraining for Generalizable Robot Controlcited by 30
  13. 2026CLAP: Contrastive Latent Action Pretraining for Learning Vision-Language-Action Models from Human Videoscited by 27
  14. 2026When Vision Overrides Language: Evaluating and Mitigating Counterfactual Failures in VLAscited by 27
  15. 2026StarVLA-α: Reducing Complexity in Vision-Language-Action Systemscited by 27
  16. 2026QuantVLA: Scale-Calibrated Post-Training Quantization for Vision-Language-Action Modelscited by 25
  17. 2026RLDX-1 Technical Reportcited by 25
  18. 2026OA-WAM: Object-Addressable World Action Model for Robust Robot Manipulationcited by 25
  19. 2026LangForce: Bayesian Decomposition of Vision Language Action Models via Latent Action Queriescited by 24
  20. 2026Adaptive Action Chunking at Inference-time for Vision-Language-Action Modelscited by 24
  21. 2026Xiaomi-Robotics-1: Scaling Vision-Language-Action Models with over 100K Hours of Real-World Trajectoriescited by 24
  22. 2025In-N-On: Scaling Egocentric Manipulation with in-the-wild and on-task Datacited by 21
  23. 2026AtomVLA: Scalable Post-Training for Robotic Manipulation via Predictive Latent World Modelscited by 21
  24. 2025Mixture of Horizons in Action Chunkingcited by 20
  25. 2026HoloBrain-0 Technical Reportcited by 20
  26. 2026FASTER: Rethinking Real-Time Flow VLAscited by 20
  27. 2026Light-WAM: Efficient World Action Models with State-Fusion Action Decodingcited by 19
  28. 2026EventVLA: Event-Driven Visual Evidence Memory for Long-Horizon Vision-Language-Action Policiescited by 18
  29. 2026LIBERO-X: Robustness Litmus for Vision-Language-Action Modelscited by 17
  30. 2026LAP: Language-Action Pre-Training Enables Zero-shot Cross-Embodiment Transfercited by 17
  31. 2025AsyncVLA: Asynchronous Flow Matching for Vision-Language-Action Modelscited by 16
  32. 2026Flash-WAM: Modality-Aware Distillation for World Action Modelscited by 16
  33. 2025RobotArena ∞: Scalable Robot Benchmarking via Real-to-Sim Translationcited by 15
  34. 2025XR-1: Towards Versatile Vision-Language-Action Models via Learning Unified Vision-Motion Representationscited by 15
  35. 2026TwinBrainVLA: Unleashing the Potential of Generalist VLMs for Embodied Tasks via Asymmetric Mixture-of-Transformerscited by 14
  36. 2026Not All Features Are Created Equal: A Mechanistic Study of Vision-Language-Action Modelscited by 14
  37. 2026Next Forcing: Causal World Modeling with Multi-Chunk Predictioncited by 14
  38. 2026VLA-Corrector: Lightweight Detect-and-Correct Inference for Adaptive Action Horizoncited by 14

50 most cited of 285 All 285 (JSON)

Similar, not cited

Close in wording, no citation either way

  1. 2025VLA-Adapter: An Effective Paradigm for Tiny-Scale Vision-Language-Action Model
  2. 2025DexVLA: Vision-Language Model with Plug-In Diffusion Expert for General Robot Control
  3. 2025VLA-0: Building State-of-the-Art VLAs with Zero Modification
  4. 2026A Pragmatic VLA Foundation Model
  5. 2026VLAFlow: A Unified Training Framework for Vision-Language-Action Models via Co-training and Future Latent Alignment
  6. 2025Vision-Language-Action Models for Robotics: A Review Towards Real-World Applications
  7. 2025Evo-1: Lightweight Vision-Language-Action Model with Preserved Semantic Alignment
  8. 2026CoRE-VLA: Towards Scalable and Robust Vision-Language-Action Modeling via Conditional Routing of Experts