ScreenSpot-v-2

The 'ScreenSpot-v2' dataset/benchmark contains a collection of screenshots used to evaluate the performance of mobile device control agents in executing complex tasks requiring long-horizon planning and precise operations.

Papers using ScreenSpot-v-2 (6)

Hi-Agent: Hierarchical Vision-Language Agents for Mobile Device Control2025 · 2 citesGUI-Shift: Enhancing VLM-Based GUI Agents through Self-supervised Reinforcement Learning2025GUI-G$^2$: Gaussian Reward Modeling for GUI Grounding2025UI-AGILE: Advancing GUI Agents with Effective Reinforcement Learning and Precise Inference-Time Grounding2025K^2-Agent: Co-Evolving Know-What and Know-How for Hierarchical Mobile Device Control2026GuirlVG: Incentivize GUI Visual Grounding via Empirical Exploration on Reinforcement Learning2025