SimplerEnv

'SimplerEnv' is an 80-task benchmark designed to evaluate closed-loop control and high-level instruction understanding in vision-language-action models.

Papers using SimplerEnv (14)

TwinBrainVLA: Unleashing the Potential of Generalist VLMs for Embodied Tasks via Asymmetric Mixture-of-Transformers2026VLA-JEPA: Enhancing Vision-Language-Action Model with Latent World Model2026Spatial Traces: Enhancing VLA Models with Spatial-Temporal Understanding2025 · 2 citesPhysBrain 1.0 Technical Report2026StarVLA: A Lego-like Codebase for Vision-Language-Action Model Developing2026ReFineVLA: Multimodal Reasoning-Aware Generalist Robotic Policies via Teacher-Guided Fine-Tuning2026Efficient Long-Horizon Vision-Language-Action Models via Static-Dynamic Disentanglement2026Unified Diffusion VLA: Vision-Language-Action Model via Joint Discrete Denoising Diffusion Process2025MAPS: Preserving Vision-Language Representations via Module-Wise Proximity Scheduling for Better Vision-Language-Action Generalization2025Maniagent: An Agentic Framework For General Robotic Manipulation2025InstructVLA: Vision-Language-Action Instruction Tuning from Understanding to Manipulation2025STORM: Search-guided Generative World Models For Robotic Manipulation2025TTF-VLA: Temporal Token Fusion Via Pixel-attention Integration For Vision-language-action Models2025Self-improving Vision-language-action Models With Data Generation Via Residual RL2025