| UNITAS: A 3D-Native World Action Model for Embodied Manipulation | ✓ Link | 99.8 | UNITAS | 2026-10-08 |
| HABILIS Brain 0: Geometry-Change Supervision for Vision-Language-Action and Residual Flow Recovery | ✓ Link | 99.55 | HABILIS-Brain-0 (GC-VLA + GCRF) | 2026-09-22 |
| Long-WAM: Scaling the Context of World-Action Models | ✓ Link | 99.5 | Long-WAM (IDM) | 2026-10-07 |
| ABot-M0.5: Unified Mobility-and-Manipulation World Action Model | | 99.4 | ABot-M0.5 | 2026-07-01 |
| OpenWAM: An Open, Modular Exploration Towards Systematic World-Action Model Pretraining | ✓ Link | 99.3 | OpenWAM-alpha (extra data) | 2026-09-07 |
| SRPO: Self-Referential Policy Optimization for Vision-Language-Action Models | ✓ Link | 99.2 | Online SRPO | 2025-11-19 |
| Qwen-RobotManip Technical Report: Alignment Unlocks Scale for Robotic Manipulation Foundation Models | ✓ Link | 99.2 | Qwen-RobotManip-Context | 2026-06-17 |
| World Action Modeling with Progressive Visual Planning | | 99.2 | ProWAM-Base | 2026-10-01 |
| World Action Learning via Interaction-Centric Spectral Latent Guidance | | 99.2 | WING | 2026-10-02 |
| SimpleVLA-RL: Scaling VLA Training via Reinforcement Learning | ✓ Link | 99.1 | SimpleVLA-RL (OpenVLA-OFT) | 2025-09-11 |
| MM-ABC: Towards Generalist Mobile Manipulation via Seeing, Coordinating and Imagining | ✓ Link | 99.1 | MM-ABC | 2026-09-28 |
| Magic-W0: A Structured World-Action Foundation Model for Physical Intelligence | ✓ Link | 99.1 | Magic-W0 (extra data) | 2026-10-03 |
| Mixture of Horizons in Action Chunking | ✓ Link | 99.0 | π0.5 with MoH | 2025-11-24 |
| Native Action-Prior Learning from Videos for World Action Models | | 99.0 | NAVA-WAM | 2026-10-02 |
| RealtimeWAM: One-Step Asynchronous World Action Models | ✓ Link | 99.0 | RealtimeWAM† | 2026-10-05 |
| Being-H0.5: Scaling Human-Centric Robot Learning for Cross-Embodiment Generalization | ✓ Link | 98.9 | Being-H0.5 (specialist) | 2026-01-19 |
| InternVLA-A1.5: Unifying Understanding, Latent Foresight, and Action for Compositional Generalization | ✓ Link | 98.9 | InternVLA-A1.5-Libero | 2026-07-06 |
| StarVLA: A Lego-like Codebase for Vision-Language-Action Model Developing | ✓ Link | 98.8 | StarVLA-OFT (specialist) | 2026-04-06 |
| PhysBrain 1.0 Technical Report | ✓ Link | 98.8 | PhysBrain 1.0 | 2026-05-14 |
| Less is More: Reducing Complexity in Vision-Language-Action Systems | ✓ Link | 98.8 | StarVLA-α (Specialist) | 2026-09-08 |
| SnapFlow: One-Step Action Generation for Flow-Matching VLAs via Progressive Self-Distillation | | 98.75 | π0.5 + SnapFlow (1-step) | 2026-04-07 |
| Xiaomi-Robotics-0: An Open-Sourced Vision-Language-Action Model with Real-Time Execution | ✓ Link | 98.7 | Xiaomi-Robotics-0 | 2026-02-13 |
| ST-WAM: Semantic-Temporal World Action Model for Robust Manipulation under Visual Distribution Shifts | ✓ Link | 98.7 | ST-WAM | 2026-07-31 |
| WAVE-0: Bridging VLA Policies and WAMs with Lightweight Latent Future Prediction | ✓ Link | 98.7 | WAVE-0 | 2026-10-13 |
| When to Switch: Reliable Action-Chunk Extension for Vision-Language-Action Models | ✓ Link | 98.65 | RACE (H_exec=10) | 2026-10-05 |
| Hume: Introducing System-2 Thinking in Visual-Language-Action Model | ✓ Link | 98.6 | Hume | 2025-05-27 |
| ABot-M0: VLA Foundation Model for Robotic Manipulation with Action Manifold Learning | ✓ Link | 98.6 | ABot-M0 | 2026-02-11 |
| SimVLA: A Simple VLA Baseline for Robotic Manipulation | ✓ Link | 98.6 | SimVLA | 2026-02-20 |
| IntentVLA: Short-Horizon Intent Modeling for Aliased Robot Manipulation | ✓ Link | 98.6 | IntentVLA | 2026-05-14 |
| FM-Steer: Enhance Generalist Policies with Value-Guided Cascaded Denoising | | 98.6 | FM-Steer (π0) | 2026-06-03 |
| LaWAM: Latent World Action Models for Efficient Dynamics-Aware Robot Policies | ✓ Link | 98.6 | LaWAM | 2026-06-14 |
| Learning Transferable Dynamics Priors from Action to World Modeling | | 98.6 | A2World-policy | 2026-06-28 |
| InternW0: A Foundational Physical World Model for Efficient Real-World Interactions | | 98.6 | InternW0 (extra data) | 2026-09-23 |
| TacitVLA: Reasoning Vision-Language-Action Models Without Reasoning Latency | | 98.6 | TacitVLA | 2026-10-13 |
| MorphAct: Robot Policies Morphing with the Worlds They Act In | ✓ Link | 98.6 | MorphAct (pi0.5) | 2026-10-14 |
| VLA-Adapter: An Effective Paradigm for Tiny-Scale Vision-Language-Action Model | ✓ Link | 98.5 | VLA-Adapter-Pro | 2025-09-11 |
| Spatial Forcing: Implicit Spatial Representation Alignment for Vision-language-action Model | ✓ Link | 98.5 | Spatial Forcing | 2025-10-14 |
| ACoT-VLA: Action Chain-of-Thought for Vision-Language-Action Models | ✓ Link | 98.5 | ACoT-VLA | 2026-01-16 |
| Cosmos Policy: Fine-Tuning Video Models for Visuomotor Control and Planning | | 98.5 | Cosmos Policy | 2026-01-22 |
| Causal World Modeling for Robot Control | ✓ Link | 98.5 | LingBot-VA | 2026-01-29 |
| Fast-WAM: Do World Action Models Need Test-time Future Imagination? | ✓ Link | 98.5 | Fast-WAM-Joint | 2026-03-17 |
| World Pilot: Steering Vision-Language-Action Models with World-Action Priors | ✓ Link | 98.5 | World Pilot | 2026-06-10 |
| SG-WAM: Self-Guided World Modeling in Geometry-Aware Policy Space | ✓ Link | 98.5 | SG-WAM (0.9B) | 2026-08-02 |
| Vision-Language-Delta-Action: Learning Compact Deltas as Actionable Transition Interfaces for Robotic Manipulation | ✓ Link | 98.5 | LoongForge-VLDA | 2026-10-19 |
| BayesianVLA: Bayesian Decomposition of Vision Language Action Models via Latent Action Queries | ✓ Link | 98.4 | LangForce | 2026-01-21 |
| ImageWAM: Do World Action Models Really Need Video Generation, or Just Image Editing? | ✓ Link | 98.4 | ImageWAM (FLUX.2 4B) | 2026-06-17 |
| Breaking the Vision-Action Shortcut: Latent Interface Training for Generalizable Robotics Foundation Models | ✓ Link | 98.4 | ImageWAM + LIT | 2026-09-11 |
| Beyond Future Prediction: Denoising as Generative Adaptation for Robot Control | ✓ Link | 98.4 | NowWAM (FLUX.2-Klein-4B) | 2026-09-23 |
| SimForcing: Distilling Simulation Motion Priors into Real-Domain Robot World Models | ✓ Link | 98.4 | SimForcing (Our WM) | 2026-10-05 |
| PAIR: Bridging Perception and Action in Vision-Language-Action Models | | 98.4 | VLA-Adapter + PAIR | 2026-10-06 |
| VLA-Δ: Learning Physical State Transition for Vision-Language-Action Models | ✓ Link | 98.4 | VLA-Δ | 2026-10-15 |
| DiG-Flow: Discrepancy-Guided Flow Matching for Robust VLA Models | ✓ Link | 98.3 | π0.5 + DiG | 2025-12-01 |
| Mimic Intent, Not Just Trajectories | ✓ Link | 98.3 | MINT-4B | 2026-03-28 |
| Drop-Then-Recovery: How Redundant Are Vision-Language-Action Models? | ✓ Link | 98.3 | OpenVLA-OFT + DTR (Drop Half Language) | 2026-06-26 |
| StreamPI: Streaming Multimodal Temporal Modeling for Vision-Language-Action Models | ✓ Link | 98.3 | StreamPI (T=5) | 2026-08-26 |
| LaMP: Learning Vision-Language-Action Policies with 3D Scene Flow as Latent Motion Prior | ✓ Link | 98.3 | LaMP | 2026-09-08 |
| PINT: A Unified Model with Shared Predictive Intent for Robot Manipulation | ✓ Link | 98.2 | PINT-VA | 2026-10-12 |
| X-VLA: Soft-Prompted Transformer as Scalable Cross-Embodiment Vision-Language-Action Model | ✓ Link | 98.1 | X-VLA-0.9B | 2025-10-11 |
| MolmoAct2: Action Reasoning Models for Real-world Deployment | ✓ Link | 98.10 | MolmoAct2-Think | 2026-05-04 |
| Dynin-Robotics: Omnimodal Unified Diffusion Vision-Language-Action Model | ✓ Link | 98.1 | Dynin-Robotics | 2026-09-11 |
| Rolling-WAM: World Action Models with Rolling Imagination | ✓ Link | 98.1 | Rolling-WAM | 2026-09-24 |
| HiF-VLA: Hindsight, Insight and Foresight through Motion Representation for Vision-Language-Action Models | ✓ Link | 98.0 | HiF-VLA | 2025-12-10 |
| MMaDA-VLA: Large Diffusion Vision-Language-Action Model with Unified Multi-Modal Instruction and Generation | ✓ Link | 98.0 | MMaDA-VLA | 2026-03-26 |
| Qwen-VLA: Unifying Vision-Language-Action Modeling across Tasks, Environments, and Robot Embodiments | ✓ Link | 97.9 | Qwen-VLA-Instruct | 2026-05-28 |
| DRIFT: A Residual Flow Adapter for Decoding Continuous Outputs in Vision-Language Models | ✓ Link | 97.9 | DRIFT (Qwen3-VL-2B) | 2026-06-04 |
| Think Like a World Model, Act Like a VLA: Distilling World-Model Representations into Compact Robot Policies | ✓ Link | 97.9 | THAW-VLA (0.8B) | 2026-09-21 |
| Kinematic MeanFlow: One-Step Action Generation Policy for Robotic Foundation Models | ✓ Link | 97.9 | K-MF (GR00T-N1.6) | 2026-10-01 |
| Conservative Offline Robot Policy Learning via Posterior-Transition Reweighting | ✓ Link | 97.8 | PTR | 2026-03-17 |
| RLDX-1 Technical Report | ✓ Link | 97.8 | RLDX-1 (extra data) | 2026-05-05 |
| VLA-Corrector: Lightweight Detect-and-Correct Inference for Adaptive Action Horizon | ✓ Link | 97.8 | π0.5 (Few-shot) + VLA-Corrector | 2026-07-02 |
| Enfold: Folding World Model Imagination into Predictive Representations for Ultra-Efficient Embodied Control | ✓ Link | 97.8 | Enfold | 2026-08-06 |
| GeoVLA: Empowering 3D Representations in Vision-Language-Action Models | | 97.7 | GeoVLA | 2025-08-12 |
| HAMLET: Switch your Vision-Language-Action Model into a History-Aware Policy | | 97.7 | HAMLET | 2025-10-01 |
| TBD-VLA: Temporal Block Diffusion Vision Language Action Model | | 97.7 | TBD-VLA | 2026-06-05 |
| TurboVLA: Real-Time Vision-Language-Action Model at 32 Hz on an RTX 4090 with <1 GB VRAM | ✓ Link | 97.7 | TurboVLA (DINOv3 ViT-B) | 2026-07-29 |
| Being-H0.5: Scaling Human-Centric Robot Learning for Cross-Embodiment Generalization | ✓ Link | 97.6 | Being-H0.5 (generalist) | 2026-01-19 |
| Fast-WAM: Do World Action Models Need Test-time Future Imagination? | ✓ Link | 97.6 | Fast-WAM | 2026-03-17 |
| Geometric Action Model for Robot Policy Learning | ✓ Link | 97.6 | GAM | 2026-06-15 |
| REKI: Registering Knowledge in VLAs | ✓ Link | 97.6 | REKI | 2026-10-15 |
| Interactive Post-Training for Vision-Language-Action Models | ✓ Link | 97.5 | OpenVLA-OFT + RIPT | 2025-05-22 |
| CogVLA: Cognition-Aligned Vision-Language-Action Model via Instruction-Driven Routing & Sparsification | ✓ Link | 97.4 | CogVLA | 2025-08-28 |
| The Better You Learn, the Smarter You Prune: Towards Efficient Vision-Language-Action Models Via Differentiable Token Pruning | | 97.4 | LightVLA (LIBERO average) | 2025-09-16 |
| RynnVLA-002: A Unified Vision-Language-Action and World Model | ✓ Link | 97.4 | RynnVLA-002 | 2025-11-21 |
| VLANeXt: Recipes for Building Strong VLA Models | ✓ Link | 97.4 | VLANeXt | 2026-02-20 |
| FrameSkip: Learning from Fewer but More Informative Frames in VLA Training | ✓ Link | 97.4 | FrameSkip | 2026-05-13 |
| EmbodiedSkills: A Unified Framework for Orchestrating, Training, and Deploying VLA Agents | ✓ Link | 97.4 | EmbodiedSkills task-adapted pi0.5 | 2026-09-01 |
| GreenWAM: A World Foundation Model for Robotics | ✓ Link | 97.4 | GreenWAM | 2026-10-13 |
| CoDA: Co-Evolution of Data Engines and Action Policies | | 97.4 | CoDA (pi_0.5), n = 15 rounds (extra data) | 2026-10-15 |
| Generation Models Know Space: Unleashing Implicit 3D Priors for Scene Understanding | ✓ Link | 97.3 | VEGA-3D | 2026-03-19 |
| Embodied-R1.5: Evolving Physical Intelligence via Embodied Foundation Models | ✓ Link | 97.3 | Embodied-R1.5-VLA | 2026-06-09 |
| Rethinking Muon Beyond Pretraining: Spectral Failures and High-Pass Remedies for VLA and RLVR | ✓ Link | 97.25 | Pion (default, VLA-Adapter) | 2026-05-19 |
| Dream-VL & Dream-VLA: Open Vision-Language and Vision-Language-Action Models with Diffusion Language Model Backbone | ✓ Link | 97.2 | Dream-VLA | 2025-12-27 |
| VLA-JEPA: Enhancing Vision-Language-Action Model with Latent World Model | ✓ Link | 97.2 | VLA-JEPA | 2026-02-10 |
| LAP: Language-Action Pre-Training Enables Zero-shot Cross-Embodiment Transfer | | 97.2 | LAP-3B + VQA Co-training | 2026-02-11 |
| MolmoAct2: Action Reasoning Models for Real-world Deployment | ✓ Link | 97.20 | MolmoAct2 | 2026-05-04 |
| Light-WAM: Efficient World Action Models with State-Fusion Action Decoding | ✓ Link | 97.2 | Light-WAM | 2026-06-06 |
| Learning the 4D World through Point Trajectory Forecasting | ✓ Link | 97.2 | VPM (V+T+A) (extra data) | 2026-10-20 |
| Fine-Tuning Vision-Language-Action Models: Optimizing Speed and Success | ✓ Link | 97.1 | OpenVLA-OFT (additional inputs) | 2025-02-27 |
| FASTER: Rethinking Real-Time Flow VLAs | ✓ Link | 97.1 | X-VLA + FASTER | 2026-03-19 |
| StreamingVLA: Streaming Vision-Language-Action Model with Action Flow Matching and Adaptive Early Observation | | 97.1 | StreamingVLA (AFM) | 2026-03-30 |
| CapVector: Learning Transferable Capability Vectors in Parametric Space for Vision-Language-Action Models | ✓ Link | 97.1 | CapVector (with orthogonal regularization) | 2026-05-11 |
| LEAP-VLA: Latent-Enhanced Action Prototyping via Continuous Residual Latent Spaces for Vision-Language-Action Models | | 97.1 | LEAP-VLA | 2026-09-08 |
| DeepThinkVLA: Enhancing Reasoning Capability of Vision-Language-Action Models | ✓ Link | 97.0 | DeepThinkVLA (pi0-FAST) | 2025-10-31 |
| Shallow-π: Knowledge Distillation for Flow-based VLAs | ✓ Link | 97.0 | π0.5-L9 | 2026-01-28 |
| SemanticVLA: Towards Semantic Reasoning over Action Memorization via Synergistic Explicit Trace and Latent Action Planning | | 97.0 | SemanticVLA | 2026-06-03 |
| GR00T N1: An Open Foundation Model for Generalist Humanoid Robots | ✓ Link | 96.99 | GR00T N1.7 | 2025-03-18 |
| FLOWER: Democratizing Generalist Robot Policies with Efficient Vision-Language-Action Flow Policies | ✓ Link | 96.9 | FLOWER | 2025-09-05 |
| Action-Sketcher: From Reasoning to Action via Visual Sketches for Robotic Manipulation | ✓ Link | 96.9 | Action-Sketcher (extra data) | 2026-01-04 |
| StaMo: Unsupervised Learning of Generalizable Robot Motion from Compact State Representation | ✓ Link | 96.8 | OpenVLA-OFT + StaMo state | 2025-10-06 |
| Spatial-Aware VLA Pretraining through Visual-Physical Alignment from Human Videos | ✓ Link | 96.8 | VIPA-VLA | 2025-12-15 |
| Mantis: A Versatile Vision-Language-Action Model with Disentangled Visual Foresight | ✓ Link | 96.7 | Mantis | 2025-11-20 |
| FOCA: Future-Oriented Conditioning for Data-Efficient Vision-Language-Action Adaptation | | 96.7 | FOCA | 2026-06-18 |
| MemoryVLA: Perceptual-Cognitive Memory in Vision-Language-Action Models for Robotic Manipulation | | 96.65 | MemoryVLA | 2025-08-26 |
| Fast-dVLA: Accelerating Discrete Diffusion VLA to Real-Time Performance | | 96.6 | Fast-dVLA | 2026-03-26 |
| RhinoVLA Technical Report | ✓ Link | 96.6 | RhinoVLA + VR Pretrain | 2026-06-05 |
| Discrete Diffusion VLA: Bringing Discrete Diffusion to Action Decoding in Vision-Language-Action Policies | ✓ Link | 96.4 | Discrete Diffusion VLA | 2025-08-27 |
| MM-ACT: Learn from Multimodal Parallel Generation to Act | ✓ Link | 96.3 | MM-ACT (+Text in Long) | 2025-11-30 |
| GEM: Generative Supervision Helps Embodied Intelligence | ✓ Link | 96.1 | GEM-VLA | 2026-05-27 |
| APT: Action Expert Pretraining Improves Instruction Generalization of Vision-Language-Action Policies | ✓ Link | 96.1 | APT | 2026-06-10 |
| BitVLA: 1-bit Vision-Language-Action Models for Robotics Manipulation | ✓ Link | 96.0 | BitVLA | 2025-06-09 |
| InternVLA-M1: A Spatially Guided Vision-Language-Action Framework for Generalist Robot Policy | ✓ Link | 95.9 | InternVLA-M1 | 2025-10-15 |
| InstructVLA: Vision-Language-Action Instruction Tuning from Understanding to Manipulation | ✓ Link | 95.8 | InstructVLA-1.5B | 2025-07-23 |
| EVOLVE-VLA: Test-Time Training from Environment Feedback for Vision-Language-Action Models | ✓ Link | 95.8 | EVOLVE-VLA | 2025-12-16 |
| AffordanceVLA: A Vision-Language-Action Model Empowering Action Generation through Affordance-Aware Understanding | ✓ Link | 95.8 | AffordanceVLA (full) | 2026-06-04 |
| Taming VLAs under Robot Execution Errors: Self-Compensation and Stress Testing | | 95.8 | Self-Compensating VLA (pi_0.5) | 2026-09-29 |
| F1: A Vision-Language-Action Model Bridging Understanding and Generation to Actions | ✓ Link | 95.7 | F1 (pretrained) | 2025-09-08 |
| Flash-WAM: Modality-Aware Distillation for World Action Models | ✓ Link | 95.7 | Flash-WAM (1v/2a) | 2026-06-03 |
| Chain of World: World Model Thinking in Latent Motion | ✓ Link | 95.6 | Chain of World | 2026-03-03 |
| Unified Vision-Language-Action Model | ✓ Link | 95.5 | UniVLA | 2025-06-24 |
| UniVLA: Learning to Act Anywhere with Task-centric Latent Actions | ✓ Link | 95.2 | UniVLA (Task-centric Latent Actions) | 2025-05-09 |
| SwiftVLA: Unlocking Spatiotemporal Dynamics for Lightweight VLA Models at Minimal Overhead | ✓ Link | 95.1 | SwiftVLA with 4D input | 2025-11-30 |
| MemBodied: Recurrent Associative Memory for Vision-Language-Action Models | ✓ Link | 95.1 | MemBodied | 2026-09-23 |
| NORA-1.5: A Vision-Language-Action Model Trained using World Model- and Action-based Preference Rewards | ✓ Link | 95.0 | NORA-1.5 (DPO) | 2025-11-18 |
| Can World and Action Modeling Benefit Each Other under Unified Causal Training? | ✓ Link | 94.95 | OneWM | 2026-10-15 |
| DepthVLA: Enhancing Vision-Language-Action Models with Depth-Aware Spatial Reasoning | | 94.9 | DepthVLA | 2025-10-15 |
| Evo-1: Lightweight Vision-Language-Action Model with Preserved Semantic Alignment | ✓ Link | 94.8 | Evo-1 (LIBERO average) | 2025-11-06 |
| Beyond Dense Futures: World Models as Structured Planners for Robotic Manipulation | ✓ Link | 94.8 | StructVLA | 2026-03-13 |
| ActionPiece: Rethinking Action Tokenization for Autoregressive Vision-Language-Action Models | ✓ Link | 94.8 | ActionPiece + Qwen3-VL-4B | 2026-09-16 |
| Accelerating Vision-Language-Action Model Integrated with Action Chunking via Parallel Decoding | | 94.7 | PD-VLA | 2025-03-04 |
| VLA-0: Building State-of-the-Art VLAs with Zero Modification | ✓ Link | 94.7 | VLA-0 | 2025-10-15 |
| Towards More Efficient Decoding for Autoregressive Vision-language-action Models | | 94.4 | CEED-VLA (UniVLA) | 2026-09-08 |
| Knowledge Insulating Vision-Language-Action Models: Train Fast, Run Fast, Generalize Better | | 94.3 | Knowledge-Insulated VLA (from generalist model) | 2025-05-29 |
| π-StepNFT: Wider Space Needs Finer Steps in Online RL for Flow-based VLAs | ✓ Link | 94.0 | π-StepNFT (π0.5) | 2026-03-02 |
| A GPU-Parallel Framework for Heterogeneous Multi-Task Reinforcement Learning | ✓ Link | 93.5 | Vis IW-ABC | 2026-10-06 |
| SCALE: Self-uncertainty Conditioned Adaptive Looking and Execution for Vision-Language-Action Models | ✓ Link | 93.0 | π0-FAST + SCALE | 2026-02-04 |
| Recurrent-Depth VLA: Implicit Test-Time Compute Scaling of Vision-Language-Action Models via Latent Iterative Reasoning | ✓ Link | 93.0 | RD-VLA (Fixed) | 2026-02-08 |
| Unified Diffusion VLA: Vision-Language-Action Model via Joint Discrete Denoising Diffusion Process | ✓ Link | 92.7 | UD-VLA | 2025-11-03 |
| DreamVLA: A Vision-Language-Action Model Dreamed with Comprehensive World Knowledge | ✓ Link | 92.6 | DreamVLA | 2025-07-06 |
| Dita: Scaling Diffusion Transformer for Generalist Vision-Language-Action Policy | ✓ Link | 92.3 | Dita w/ wrist | 2025-03-25 |
| VLA-RFT: Vision-Language-Action Reinforcement Fine-tuning with Verified Rewards in World Simulators | ✓ Link | 91.1 | VLA-RFT | 2025-10-01 |
| CorridorVLA: Explicit Spatial Constraints for Generative Action Heads via Sparse Anchors | | 90.95 | SmolVLA-Corr | 2026-04-23 |
| CLAP: Direct VLM-to-VLA Adaptation via Language-Action Grounding | ✓ Link | 90.8 | CLAP Qwen3.5 2B | 2026-07-09 |
| E-TTS: A New Embodied Test-Time Scaling Framework for Robotic Manipulation | | 90.78 | MolmoAct + E-TTS | 2026-06-25 |
| villa-X: Enhancing Latent Action Modeling in Vision-Language-Action Models | ✓ Link | 90.1 | villa-X | 2025-07-31 |
| Reinforcing Action Policies by Prophesying | ✓ Link | 90.1 | VLA-Adapter + FA-GRPO & FlowScale | 2025-11-25 |
| Fast-ThinkAct: Efficient Vision-Language-Action Reasoning via Verbalizable Latent Planning | | 89.7 | Fast-ThinkAct (3B) | 2026-01-14 |
| SmolVLA: A Vision-Language-Action Model for Affordable and Efficient Robotics | ✓ Link | 88.75 | SmolVLA (2.25B) | 2025-06-02 |
| 4D-VLA: Spatiotemporal Vision-Language-Action Pretraining with Cross-Scene Calibration | | 88.6 | 4D-VLA | 2025-06-27 |
| G^3VLA: Geometric Inductive Bias for Vision-Language-Action Models | | 88.1 | G^3VLA (GT) | 2026-10-14 |
| NORA: A Small Open-Sourced Generalist Vision Language Action Model for Embodied Tasks | ✓ Link | 87.9 | NORA-Long-fine-tuned | 2025-04-28 |
| MetricAnything: Scaling Metric Depth Pretraining with Noisy Heterogeneous Sources | ✓ Link | 87.7 | MetricAnything (VLA depth distillation) | 2026-01-29 |
| RoboAlign: Learning Test-Time Reasoning for Language-Action Alignment in Vision-Language-Action Models | | 86.8 | RoboAlign | 2026-03-22 |
| MolmoAct: Action Reasoning Models that can Reason in Space | ✓ Link | 86.6 | MolmoAct-7B-D (extra data) | 2025-08-11 |
| FAST: Efficient Action Tokenization for Vision-Language-Action Models | ✓ Link | 85 | π0-FAST | 2025-01-16 |
| ThinkAct: Vision-Language-Action Reasoning via Reinforced Visual Latent Planning | | 84.4 | ThinkAct | 2025-07-22 |
| Diffusion Transformer Policy | ✓ Link | 82.4 | DiT Policy | 2024-10-21 |
| Dita: Scaling Diffusion Transformer for Generalist Vision-Language-Action Policy | ✓ Link | 82.4 | Dita (third-person RGB) | 2025-03-25 |
| WorldVLA: Towards Autoregressive Action World Model | ✓ Link | 81.8 | WorldVLA (512x512) | 2025-06-26 |
| FastOPD: On-Policy Distillation for Lightweight VLA Deployment | | 81.8 | FastOPD (SmolVLA, 2 NFE) | 2026-10-02 |
| CoT-VLA: Visual Chain-of-Thought Reasoning for Vision-Language-Action Models | | 81.13 | CoT-VLA-7B | 2025-03-27 |
| VLA-RL: Towards Masterful and General Robotic Manipulation with Scalable Reinforcement Learning | ✓ Link | 81.0 | VLA-RL | 2025-05-24 |
| Domain Arithmetic: One-Shot VLA Adaptation under Environmental Shifts | ✓ Link | 79.4 | pi0-FAST + DART | 2026-07-01 |
| OpenVLA: An Open-Source Vision-Language-Action Model | ✓ Link | 76.5 | OpenVLA | 2024-06-13 |
| TraceVLA: Visual Trace Prompting Enhances Spatial-Temporal Awareness for Generalist Robotic Policies | | 74.8 | TraceVLA finetuned | 2024-12-13 |
| Visual Spatial Tuning | ✓ Link | 70.8 | VST-tuned Qwen2.5-VL-3B (10K) | 2025-11-07 |