Jul 2026
Alignment Is Local: A Paired Diagnostic for GUI Agents under User-Side Persuasion
It is shown that a one-line guardrail achieves large single-shot ASR reductions, up to roughly 40 points, at near-zero over-refusal cost, which overstates deployed robustness by a systematic and predictable margin.
Haoxin An, Yunpeng Song, Zihao Bai et al.
· arXiv.org · 0 citations