BeClaude
Research2026-05-11

ProactiveMobile: A Comprehensive Benchmark for Boosting Proactive Intelligence on Mobile Devices

Source: Arxiv CS.AI

arXiv:2602.21858v4 Announce Type: replace Abstract: Multimodal large language models (MLLMs) have made significant progress in mobile agent development, yet their capabilities are predominantly confined to a reactive paradigm, where they merely execute explicit user commands. The emerging paradigm...

arxivpapersbenchmark