VLABench

VLABench is a benchmark dataset used to evaluate the performance of learned image compression methods specifically for vision-language-action models in robotic applications.

Papers using VLABench (3)

Learned Image Compression for Vision-Language-Action Models2026TAG: Target-Agnostic Guidance for Stable Object-Centric Inference in Vision-Language-Action Models2026E0: Enhancing Generalization and Fine-Grained Control in VLA Models via Tweedie Discrete Diffusion2025