mlsanityData to 7 Oct 2026

GeoVLA: Empowering 3D Representations in Vision-Language-Action Models

arXiv 2508.09071 ↗First posted 12 Aug 2025cs.ROcited by 68

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

Abstract (arXiv)

Vision-Language-Action (VLA) models have emerged as a promising approach for enabling robots to follow language instructions and predict corresponding actions. However, current VLA models mainly rely on 2D visual inputs, neglecting the rich geometric information in the 3D physical world, which limits their spatial awareness and adaptability. In this paper, we present GeoVLA, a novel VLA framework that effectively integrates 3D information to advance robotic manipulation. It uses a vision-language model (VLM) to process images and language instructions,extracting fused vision-language embeddings. In parallel, it converts depth maps into point clouds and employs a customized point encoder, called Point Embedding Network, to generate 3D geometric embeddings independently. These produced embeddings are then concatenated and processed by our proposed spatial-aware action expert, called 3D-enhanced Action Expert, which combines information from different sensor modalities to produce precise action sequences. Through extensive experiments in both simulation and real-world environments, GeoVLA demonstrates superior performance and robustness. It achieves state-of-the-art results in the LIBERO and ManiSkill2 simulation benchmarks and shows remarkable robustness in real-world tasks requiring height adaptability, scale awareness and viewpoint invariance.

Key facts. Cited by 68 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 42 of its 45 references that are in the map. First posted on arXiv 12 Aug 2025.

Builds on

42 of 45 references.

Closest ancestors · 4

Same direction, newest first

  1. 20253D CAVLA: Leveraging Depth and 3D Context to Generalize Vision Language Action Models for Unseen Tasks9 May
  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 · 38

  1. 2020Denoising Diffusion Probabilistic Modelscited by 20,851
  2. 2020Denoising Diffusion Implicit Modelscited by 9,655
  3. 2023Visual Instruction Tuningcited by 7,792
  4. 2022Scalable Diffusion Models with Transformerscited by 5,931
  5. 2023Adding Conditional Control to Text-to-Image Diffusion Modelscited by 5,466
  6. 2022Flow Matching for Generative Modelingcited by 5,426
  7. 2016PointNet: Deep Learning on Point Sets for 3D Classification and Segmentationcited by 5,223
  8. 2021RoFormer: Enhanced Transformer with Rotary Position Embeddingcited by 4,581
  9. 2022Flamingo: a Visual Language Model for Few-Shot Learningcited by 4,337
  10. 2023Qwen Technical Reportcited by 3,273
  11. 2023RT-2: Vision-Language-Action Models Transfer Web Knowledge to Robotic Controlcited by 2,993
  12. 2024OpenVLA: An Open-Source Vision-Language-Action Modelcited by 2,777
26 more
  1. 2024π0: A Vision-Language-Action Flow Model for General Robot Controlcited by 2,297
  2. 2022RT-1: Robotics Transformer for Real-World Control at Scalecited by 2,007
  3. 2023Learning Fine-Grained Bimanual Manipulation with Low-Cost Hardwarecited by 1,490
  4. 2024Octo: An Open-Source Generalist Robot Policycited by 1,427
  5. 2023LIBERO: Benchmarking Knowledge Transfer for Lifelong Robot Learningcited by 1,338
  6. 2025GR00T N1: An Open Foundation Model for Generalist Humanoid Robotscited by 1,167
  7. 2017Mixed Precision Trainingcited by 1,135
  8. 2025Fine-Tuning Vision-Language-Action Models: Optimizing Speed and Successcited by 844
  9. 2017PoseCNN: A Convolutional Neural Network for 6D Object Pose Estimation in Cluttered Scenescited by 768
  10. 2022PaLI: A Jointly-Scaled Multilingual Language-Image Modelcited by 762
  11. 2024RDT-1B: a Diffusion Foundation Model for Bimanual Manipulationcited by 670
  12. 2025FAST: Efficient Action Tokenization for Vision-Language-Action Modelscited by 643
  13. 2022Perceiver-Actor: A Multi-Task Transformer for Robotic Manipulationcited by 545
  14. 2024CogACT: A Foundational Vision-Language-Action Model for Synergizing Cognition and Action in Robotic Manipulationcited by 411
  15. 2024Prismatic VLMs: Investigating the Design Space of Visually-Conditioned Language Modelscited by 336
  16. 2021ManiSkill: Generalizable Manipulation Skill Benchmark with Large-Scale Demonstrationscited by 193
  17. 2024LLaVA-3D: A Simple yet Effective Pathway to Empowering LMMs with 3D-awarenesscited by 178
  18. 2025HybridVLA: Collaborative Diffusion and Autoregression in a Unified Vision-Language-Action Modelcited by 176
  19. 2023EmbodiedScan: A Holistic Multi-Modal 3D Perception Suite Towards Embodied AIcited by 146
  20. 2024RVT-2: Learning Precise Manipulation from Few Demonstrationscited by 135
  21. 2022Autoencoders as Cross-Modal Teachers: Can Pretrained 2D Image Transformers Help 3D Representation Learning?cited by 121
  22. 2024LLARVA: Vision-Action Instruction Tuning Enhances Robot Learningcited by 87
  23. 2025Dita: Scaling Diffusion Transformer for Generalist Vision-Language-Action Policycited by 73
  24. 2024Lift3D Foundation Policy: Lifting 2D Large-Scale Pretrained Models for Robust 3D Robotic Manipulationcited by 45
  25. 2025DenseGrounding: Improving Dense Language-Vision Semantics for Ego-Centric 3D Visual Groundingcited by 12
  26. 2025DEGround: An Effective Baseline for Ego-centric 3D Visual Grounding with a Homogeneous Frameworkcited by 4

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 13; 2026: 55.

68 papers.

Sort:

By area:

  1. 2025MemoryVLA: Perceptual-Cognitive Memory in Vision-Language-Action Models for Robotic Manipulationcited by 183
  2. 2025LIBERO-PRO: Towards Robust and Fair Evaluation of Vision-Language-Action Models Beyond Memorizationcited by 166
  3. 2025Spatial Forcing: Implicit Spatial Representation Alignment for Vision-language-action Modelcited by 125
  4. 2026A Pragmatic VLA Foundation Modelcited by 97
  5. 2026Unified 4D World Action Modeling from Video Priors with Asynchronous Denoisingcited by 65
  6. 2025The Better You Learn, The Smarter You Prune: Towards Efficient Vision-language-action Models via Differentiable Token Pruningcited by 34
  7. 2025From Spatial to Actions: Grounding Vision-Language-Action Model in Spatial Foundation Priorscited by 25
  8. 2025SwiftVLA: Unlocking Spatiotemporal Dynamics for Lightweight VLA Models at Minimal Overheadcited by 23
  9. 2025QDepth-VLA: Quantized Depth Prediction as Auxiliary Supervision for Vision-Language-Action Modelscited by 19
  10. 2026VTAM: Video-Tactile-Action Models for Complex Physical Interaction Beyond VLAscited by 14
  11. 2026Learning 4D Geometric Priors for Inference-Efficient World Action Modelscited by 14
  12. 2026Any3D-VLA: Enhancing VLA Robustness via Diverse Point Cloudscited by 13
38 more
  1. 2025Dexbotic: Open-Source Vision-Language-Action Toolboxcited by 12
  2. 2025Affordance Field Intervention: Enabling VLAs to Escape Memory Traps in Robotic Manipulationcited by 12
  3. 2026DM0: An Embodied-Native Vision-Language-Action Model towards Physical AIcited by 12
  4. 2026WAM4D: Fast 4D World Action Model via Spatial Register Tokenscited by 12
  5. 2026ROCKET: Residual-Oriented Multi-Layer Alignment for Spatially-Aware Vision-Language-Action Modelscited by 11
  6. 2026LaMP: Learning Vision-Language-Action Policy with 3D Scene Flow as Latent Motion Priorcited by 11
  7. 2025GLaD: Geometric Latent Distillation for Vision-Language-Action Modelscited by 10
  8. 2026SimVLA: A Simple VLA Baseline for Robotic Manipulationcited by 10
  9. 2026Geometric Action Model for Robot Policy Learningcited by 10
  10. 2026MemoryVLA++: Temporal Modeling via Memory and Imagination in Vision-Language-Action Modelscited by 9
  11. 2026NovaPlan: Zero-Shot Long-Horizon Manipulation via Closed-Loop Video Language Planningcited by 7
  12. 2026Pri4R: Learning World Dynamics for Vision-Language-Action Models with Privileged 4D Representationcited by 7
  13. 2026GuidedVLA: Specifying Task-Relevant Factors via Plug-and-Play Action Attention Specializationcited by 7
  14. 2026PEAfowl: Perception-Enhanced Multi-View Vision-Language-Action for Bimanual Manipulationcited by 6
  15. 2026ConsisVLA-4D: Advancing Spatiotemporal Consistency in Efficient 3D-Perception and 4D-Reasoning for Robotic Manipulationcited by 6
  16. 2026PointACT: Vision-Language-Action Models with Multi-Scale Point-Action Interactioncited by 6
  17. 2026DreamWAM: Beyond RGB Future Prediction for World Action Modelscited by 6
  18. 2025SpatialActor: Exploring Disentangled Spatial Representations for Robust Robotic Manipulationcited by 5
  19. 2025An Anatomy of Vision-Language-Action Models: From Modules to Milestones and Challengescited by 4
  20. 2026E-VLA: Event-Augmented Vision-Language-Action Model for Dark and Blurred Scenescited by 4
  21. 2026Spatial Memory for Out-of-Vision Manipulation in Vision-Language-Actioncited by 4
  22. 2025VLA-LPAF: Lightweight Perspective-Adaptive Fusion for Vision-Language-Action to Enable More Unconstrained Robotic Manipulationcited by 3
  23. 2026GaussianDream: A Feed-Forward 3D Gaussian World Model for Robotic Manipulationcited by 3
  24. 2026VEGA: Visual Encoder Grounding Alignment for Spatially-Aware Vision-Language-Action Modelscited by 2
  25. 2026Robots Need More than VLA and World Modelscited by 2
  26. 2026MotionVLA: Injecting Geometric Motion into Vision-Language-Action Modelcited by 2
  27. 2026GIFT: Guided Intermediate Feature Training via Action-Oriented Structural Supervision for Robotic Manipulationcited by 2
  28. 2026Beyond Dense Futures: World Models as Structured Planners for Robotic Manipulationcited by 1
  29. 2026PVI: Plug-in Visual Injection for Vision-Language-Action Modelscited by 1
  30. 2026CoEnv: Driving Embodied Multi-Agent Collaboration via Compositional Environmentcited by 1
  31. 2026R3D: Revisiting 3D Policy Learningcited by 1
  32. 2026ELAN4D: Embodiment-Centric 4D Supervision for Vision-Language-Action Models via Plug-and-Play Adaptationcited by 1
  33. 2026GeoAlign: Beyond Semantics with State-Guided Spatial Alignment in VLA Modelscited by 1
  34. 20263DThinkVLA: Endowing Vision-Language-Action Models with Latent 3D Priors via 3D-Thinking-Guided Co-trainingcited by 1
  35. 2026See like a Robot: Robot-Centric Pointmaps for VLA Modelscited by 1
  36. 2026SAM3D-Guided Object-Centric Representation Alignment for Vision-Language-Action Modelscited by 1
  37. 2026AnchorVLA4D: an Anchor-Based Spatial-Temporal Vision-Language-Action Model for Robotic Manipulationcited by 0
  38. 2026SpatialPoint: Spatial-aware Point Prediction for Embodied Localizationcited by 0

50 most cited of 68 All 68 (JSON)

Similar, not cited

Close in wording, no citation either way

  1. 2026AugVLA-3D: Depth-Driven Feature Augmentation for Vision-Language-Action Models
  2. 2025DepthVLA: Enhancing Vision-Language-Action Models with Depth-Aware Spatial Reasoning
  3. 2026GeoBridge-VLA: Geometry-Aware Residual Adaptation for Vision-Language-Action Models
  4. 2025Evo-0: Vision-Language-Action Model with Implicit Spatial Understanding
  5. 2026G3VLA: Geometric inductive bias for Vision-Language-Action Models
  6. 2026BridgeVLA++: A Data-Efficient, Generalizable, and Memory-Augmented Vision-Language-Action Framework for 3D Manipulation
  7. 20263DVLA: Enhancing Vision-Language-Action Models via 3D Spatial and Instance Understanding
  8. 2025Spatial-Aware VLA Pretraining through Visual-Physical Alignment from Human Videos