Reasoning for VLAs
VLA: Vision-Language-Action Model
Overview
This area studies how vision-language-action agents can interpret instructions, track spatial and physical information, plan over time, and turn decisions into robot, navigation, or driving actions. Approaches include explicit text or visual chain-of-thought, 3D spatial representations, latent plans and world models, hierarchical planners with action controllers, and training with demonstrations, teacher-generated rationales, or reinforcement learning. Evaluation spans simulated and real-robot manipulation, vision-language navigation, and autonomous driving, with growing attention to long-horizon tasks, perturbations, and instruction changes. A central open question is whether reasoning actually causes better, more faithful actions—or merely accompanies behavior driven by learned priors—while added deliberation must be balanced against latency, robustness, and reliable evaluation of physical understanding.
Latest papers
122 papers17 papers in the last four weeks, down 15% on the four weeks before. 0.1% of all new papers.
UniWAM: Unified World-Action Model
MotorMind: Scaffolding General Vision Language Models for Zero-Shot Robot Manipulation
What Makes World Action Models Generalize? An Empirical Study of Test-Time Future Modeling
Resolving State-Representation Mismatch: State-Space Visual Reasoning for Open-Loop VLA Planning
RCVLA: 4D Radar-Grounded Semantic Reasoning and Trajectory Arbitration for Autonomous Driving
CausalDriveBench: Evaluating Causal Reasoning in Vision-Language-Action Models for Autonomous Driving
AdaHVLA: Adaptive Harnesses for Long-Horizon Vision-Language-Action Execution
H-VLA: Hierarchical Vision-Language-Action Model with Key-Action Reasoning and Motion Planning in a Unified Action Space
GroundingVLN: Reasoning and Acting with Grounding for Vision-Language Navigation
Grounding Vision-Language Models in Driving Semantics: A Multi-Dataset Predicate Framework for Explainable Reasoning
MobileVLA-R1 2.0: RL-Enhanced Reasoning for Mobile Robot Control
RoboSPA: Can VLA Models Go Beyond Simple Scenes and Short-Horizon Tasks?
RoboSPA focuses on two core dimensions, Fine-Grained Spatial Reasoning and Long-Horizon Procedural Planning, covering 10 task categories and 56 base tasks. Each task is instantiated across five difficulty levels, yielding 280 variants with increasing spatial ambiguity and procedural complexity. We collect 527K trajectories across multiple embodiments and diverse scenes. Beyond binary success rate, RoboSPA introduces diagnostic metrics for more detailed evaluation. Experiments on representative VLA models show that current systems still struggle with complex spatial relations, precise low-level execution, and memory-intensive planning. These results establish RoboSPA as a challenging diagnostic benchmark for developing more capable, reliable, and generalizable embodied agents. Our data and code are available at https://github.com/fanzhenxuan/RoboSPA.