一作ACL 2026Poster· Natural Language Processing
DAC -Bench: A Decision-Aware Benchmark for Compositional Mobile GUI Tasks
Yuqing Zhang · Honghui Sheng · Xueyu Hu · Shengyu Zhang · Fei Wu
摘要
Mobile GUI agents powered by LMMs can perceive screens and follow instructions, yet existing benchmarks largely target short, linear workflows and step-level accuracy, offering limited insight into long-horizon planning and decision-making under branching structures. We present DAC-Bench, a decis...