mlsanityData to 7 Oct 2026

Evo-0: Vision-Language-Action Model with Implicit Spatial Understanding

arXiv 2507.00416 ↗First posted 1 Jul 2025cs.RO cs.CVcited by 55

On the map: Vision-language-action models → 3D and spatial-aware VLA

Abstract (arXiv)

Vision-Language-Action (VLA) models have emerged as a promising framework for enabling generalist robots capable of perceiving, reasoning, and acting in the real world. These models usually build upon pretrained Vision-Language Models (VLMs), which excel at semantic understanding due to large-scale image and text pretraining. However, existing VLMs typically lack precise spatial understanding capabilities, as they are primarily tuned on 2D image-text pairs without 3D supervision. To address this limitation, recent approaches have incorporated explicit 3D inputs such as point clouds or depth maps, but this necessitates additional depth sensors or pre-trained depth estimation models, which may yield defective results. In contrast, our work introduces a plug-and-play module that implicitly incorporates 3D geometry features into VLA models by leveraging an off-the-shelf visual geometry foundation model. This integration provides the model with depth-aware visual representations, improving its ability to understand the geometric structure of the scene and the spatial relationships among objects from RGB images alone. We evaluate our method on a set of spatially challenging tasks in both simulation and the real world. Extensive evaluations show that our method significantly improves the performance of state-of-the-art VLA models across diverse scenarios.

Key facts. Cited by 55 papers in the map (data through 7 Oct 2026). Placed in 3D and spatial-aware VLA, one of 13 directions in Vision-language-action models. Builds on 31 of its 34 references that are in the map. First posted on arXiv 1 Jul 2025.

Builds on

31 of 34 references.

Closest ancestors · 4

Same direction, newest first

  1. 2025BridgeVLA: Input-Output Alignment for Efficient 3D Manipulation Learning with Vision-Language Models9 Jun
  2. 2025PointVLA: Injecting the 3D World into Vision-Language-Action Models10 Mar
  3. 2025SpatialVLA: Exploring Spatial Representations for Visual-Language-Action Model27 Jan
  4. 20243D-VLA: A 3D Vision-Language-Action Generative World Model14 Mar

Other references · 27

  1. 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,456
  2. 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 29,915
  3. 2023LLaMA: Open and Efficient Foundation Language Modelscited by 13,958
  4. 2023Llama 2: Open Foundation and Fine-Tuned Chat Modelscited by 12,865
  5. 2022Flow Matching for Generative Modelingcited by 5,426
  6. 2023RT-2: Vision-Language-Action Models Transfer Web Knowledge to Robotic Controlcited by 2,993
  7. 2024OpenVLA: An Open-Source Vision-Language-Action Modelcited by 2,777
  8. 2024π0: A Vision-Language-Action Flow Model for General Robot Controlcited by 2,297
  9. 2023DUSt3R: Geometric 3D Vision Made Easycited by 1,526
  10. 2025VGGT: Visual Geometry Grounded Transformercited by 1,475
  11. 2024Depth Anything: Unleashing the Power of Large-Scale Unlabeled Datacited by 1,346
  12. 2025GR00T N1: An Open Foundation Model for Generalist Humanoid Robotscited by 1,167
15 more
  1. 2024Grounding Image Matching in 3D with MASt3Rcited by 960
  2. 2025Fine-Tuning Vision-Language-Action Models: Optimizing Speed and Successcited by 844
  3. 2024PaliGemma: A versatile 3B VLM for transfercited by 679
  4. 2024RDT-1B: a Diffusion Foundation Model for Bimanual Manipulationcited by 670
  5. 2023ZoeDepth: Zero-shot Transfer by Combining Relative and Metric Depthcited by 646
  6. 2022Rectified Flow: A Marginal Preserving Approach to Optimal Transportcited by 277
  7. 2024MegaSaM: Accurate, Fast, and Robust Structure and Motion from Casual Dynamic Videoscited by 224
  8. 2025Spatial-MLLM: Boosting MLLM Capabilities in Visual-based Spatial Intelligencecited by 206
  9. 2025HybridVLA: Collaborative Diffusion and Autoregression in a Unified Vision-Language-Action Modelcited by 176
  10. 2024SpatialBot: Precise Spatial Understanding with Vision Language Modelscited by 146
  11. 2024RVT-2: Learning Precise Manipulation from Few Demonstrationscited by 135
  12. 2025VGGT-SLAM: Dense RGB SLAM Optimized on the SL(4) Manifoldcited by 130
  13. 2025MM-Spatial: Exploring 3D Spatial Understanding in Multimodal LLMscited by 72
  14. 2024Lift3D Foundation Policy: Lifting 2D Large-Scale Pretrained Models for Robust 3D Robotic Manipulationcited by 45
  15. 2024SUGAR: Pre-training 3D Visual Representations for Roboticscited by 26

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 15; 2026: 40.

55 papers.

Sort:

By area:

  1. 2025Spatial Forcing: Implicit Spatial Representation Alignment for Vision-language-action Modelcited by 125
  2. 2025Vision-Language-Action Models for Robotics: A Review Towards Real-World Applicationscited by 82
  3. 2026Unified 4D World Action Modeling from Video Priors with Asynchronous Denoisingcited by 65
  4. 2025Pure Vision Language Action (VLA) Models: A Comprehensive Surveycited by 46
  5. 2025Evo-1: Lightweight Vision-Language-Action Model with Preserved Semantic Alignmentcited by 35
  6. 2025VLSA: Vision-Language-Action Models with Plug-and-Play Safety Constraint Layercited by 32
  7. 2025Why Do MLLMs Struggle with Spatial Understanding? A Systematic Analysis from Data to Architecturecited by 30
  8. 2026ACoT-VLA: Action Chain-of-Thought for Vision-Language-Action Modelscited by 30
  9. 2025From Spatial to Actions: Grounding Vision-Language-Action Model in Spatial Foundation Priorscited by 25
  10. 2025GeoAware-VLA: Implicit Geometry Aware Vision-Language-Action Modelcited by 24
  11. 2025SwiftVLA: Unlocking Spatiotemporal Dynamics for Lightweight VLA Models at Minimal Overheadcited by 23
  12. 2026JEPA-VLA: Video Predictive Embedding is Needed for VLA Modelscited by 22
38 more
  1. 2025Mixture of Horizons in Action Chunkingcited by 20
  2. 2026PALM: Progress-Aware Policy Learning via Affordance Reasoning for Long-Horizon Robotic Manipulationcited by 20
  3. 2026VGGDrive: Empowering Vision-Language Models with Cross-View Geometric Grounding for Autonomous Drivingcited by 19
  4. 2025StereoVLA: Enhancing Vision-Language-Action Models with Stereo Visioncited by 16
  5. 2025Learning to Feel the Future: DreamTacVLA for Contact-Rich Manipulationcited by 14
  6. 2026Any3D-VLA: Enhancing VLA Robustness via Diverse Point Cloudscited by 13
  7. 2026Evo-Depth: A Lightweight Depth-Enhanced Vision-Language-Action Modelcited by 13
  8. 2026Gaze2Act: Gaze-Conditioned Vision-Language-Action Policies for Interactive Robot Manipulationcited by 8
  9. 2026IntentVLA: Short-Horizon Intent Modeling for Aliased Robot Manipulationcited by 7
  10. 20263D-Mix for VLA: A Plug-and-Play Module for Integrating VGGT-based 3D Information into Vision-Language-Action Modelscited by 6
  11. 2026ConsisVLA-4D: Advancing Spatiotemporal Consistency in Efficient 3D-Perception and 4D-Reasoning for Robotic Manipulationcited by 6
  12. 2026DTP: A Simple yet Effective Distracting Token Pruning Framework for Vision-Language Action Modelscited by 5
  13. 2026UAOR: Uncertainty-aware Observation Reinjection for Vision-Language-Action Modelscited by 5
  14. 2026LA4VLA: Learning to Act without Seeing via Language-Action Pretrainingcited by 5
  15. 2025An Anatomy of Vision-Language-Action Models: From Modules to Milestones and Challengescited by 4
  16. 2025Real2Edit2Real: Generating Robotic Demonstrations via a 3D Control Interfacecited by 4
  17. 2026Understanding the Impact of Geometric Foundation Models on Vision-Language-Action Modelscited by 4
  18. 2026OpenWorldLib: A Unified Codebase and Definition of Advanced World Modelscited by 3
  19. 2026Learning Action Manifold with Multi-view Latent Priors for Robotic Manipulationcited by 3
  20. 2026Embodied3DBench: Benchmarking Low-Level Embodied Spatial Intelligence of Vision Language Modelscited by 3
  21. 2026R3DP: Real-Time 3D-Aware Policy for Embodied Manipulationcited by 2
  22. 2026VEGA: Visual Encoder Grounding Alignment for Spatially-Aware Vision-Language-Action Modelscited by 2
  23. 2026Trust It or Not: Evidential Uncertainty for Feed-Forward 3D Reconstruction with Trust3Rcited by 2
  24. 2026Robust-WAM: Bridging Generative Pretraining and Semantic Foresight in World-Action Modelscited by 2
  25. 2026Modeling Cross-vision Synergy for Unified Large Vision Modelcited by 1
  26. 2026PVI: Plug-in Visual Injection for Vision-Language-Action Modelscited by 1
  27. 2026SpatialFly: Implicit 3D Prior-Guided Visual Reparameterization for Continuous UAV Vision-and-Language Navigationcited by 1
  28. 2026Action with Visual Primitivescited by 1
  29. 2026VeriSpace: Spatially Grounded Action Verification for Vision-Language-Action Modelscited by 1
  30. 2026GIVE: Grounding Human Gestures in Vision-Language-Action Modelscited by 1
  31. 2026See like a Robot: Robot-Centric Pointmaps for VLA Modelscited by 1
  32. 2025Freeze, Share, Shrink: Rethinking the Action Backbone in Diffusion Policiescited by 0
  33. 2026AnchorVLA4D: an Anchor-Based Spatial-Temporal Vision-Language-Action Model for Robotic Manipulationcited by 0
  34. 2026OmniVLA-RL: A Vision-Language-Action Model with Spatial Understanding and Online RLcited by 0
  35. 2026X-Imitator: Spatial-Aware Imitation Learning via Bidirectional Action-Pose Interactioncited by 0
  36. 2026Domain Arithmetic: One-Shot VLA Adaptation under Environmental Shiftscited by 0
  37. 2026The First EgoCross Challenge at EgoVis 2026: Cross-Domain Egocentric Video Question Answeringcited by 0
  38. 2026V-Link: Recovering Lost Visual Representations in Action DiT for Vision-Language-Action Modelscited by 0

50 most cited of 55 All 55 (JSON)

Similar, not cited

Close in wording, no citation either way

  1. 2025DepthVLA: Enhancing Vision-Language-Action Models with Depth-Aware Spatial Reasoning
  2. 2025GeoVLA: Empowering 3D Representations in Vision-Language-Action Models
  3. 2026AugVLA-3D: Depth-Driven Feature Augmentation for Vision-Language-Action Models
  4. 2025Spatial-Aware VLA Pretraining through Visual-Physical Alignment from Human Videos
  5. 2026G3VLA: Geometric inductive bias for Vision-Language-Action Models
  6. 2026GeoBridge-VLA: Geometry-Aware Residual Adaptation for Vision-Language-Action Models
  7. 2025VLA-Adapter: An Effective Paradigm for Tiny-Scale Vision-Language-Action Model
  8. 2024A3VLM: Actionable Articulation-Aware Vision Language Model