WidowX

The 'WidowX' dataset/benchmark contains a collection of tasks and scenarios designed to evaluate the performance and adaptability of Vision-Language-Action models in multi-task robotic learning.

Papers using WidowX (4)

CORAL: Scalable Multi-Task Robot Learning via LoRA Experts2026Self-Supervised Bootstrapping of Action-Predictive Embodied Reasoning2026InternVLA-M1: A Spatially Guided Vision-Language-Action Framework for Generalist Robot Policy2025Vlaser: Vision-Language-Action Model with Synergistic Embodied Reasoning2025