As personal agents evolve to handle complex, user-centric tasks, static plain-text chat is rapidly becoming a bottleneck. Generative UI emerges as the necessary new interface layer, dynamically synthesizing the right controls, options, and state from the interaction context in real time. We present Macaron-A2UI, a model for Generative UI in personal agents. Our goal is to move beyond text-only interaction by enabling agents to generate natural language together with lightweight, executable UI actions for information collection, preference refinement, confirmation, and multi-goal organization. We build a large-scale Generative UI corpus from heterogeneous dialogue sources, introduce A2UI-Bench for controlled evaluation, and train 30B, 235B and 754B models with parameter-efficient LoRA-based supervised fine-tuning followed by reward-driven reinforcement learning. The best Macaron-A2UI model reaches 75.6 overall on A2UI-Bench without explicit schema hints, surpassing the strongest full-schema frontier baseline. We release the models, benchmark, and evaluation protocol to support future work on Generative UI for personal agents.

翻译：随着个人代理在处理复杂、以用户为中心的任务时不断发展，静态纯文本聊天正迅速成为瓶颈。生成式用户界面作为一种必要的全新交互层应运而生，能够基于交互上下文实时动态合成合适的控件、选项与状态。我们提出Macaron-A2UI，一种面向个人代理的生成式UI模型。目标是突破纯文本交互的限制，使代理能够生成自然语言并同时生成轻量级、可执行的UI动作，以支持信息收集、偏好细化、确认及多目标组织。我们从异构对话源构建大规模生成式UI语料库，引入A2UI-Bench进行受控评估，并通过参数高效的基于LoRA的有监督微调，再结合奖励驱动的强化学习，训练了30B、235B和754B参数的模型。最佳Macaron-A2UI模型在无显式模式提示下，在A2UI-Bench上取得75.6的总分，超越了最强的完整模式基线。我们公开模型、基准测试及评估协议，以支持面向个人代理的生成式UI领域的未来研究。