arXiv cs.AI· Linqiang Guo (Peter), Wei Liu (Peter), Li Gu (Peter), Yang Wang (Peter), Tse-Hsun (Peter), Chen·· 11 小时前AI 评分38
Q-Reflector：基于 Qwen3-VL-8B 学习带生成条件的 GUI 动作反思
Q-Reflector: Learning GUI Action Reflection with Generated Conditions
AI 导读
Q-Reflector 是基于 Qwen3-VL-8B 的独立反思模型，将 GUI 动作反思建模为监督式结构化预测，先生成动作前置与后置条件，再给出执行成功与任务有效性的二元判定。
正文 · 原文
Abstract:Mobile GUI agents require accurate action reflection for reliable long-horizon execution, assessing whether each action was successfully executed and whether its outcome constitutes a valid step toward the task goal. When evaluated as standalone reflectors under a common prompt, frontier VLMs and specialized GUI models exhibit imbalanced recognition of valid and invalid transitions. We propose Q-Reflector, a standalone reflector based on Qwen3-VL-8B that learns GUI action reflection as supervised structured prediction. Given task context, the executed action, and paired screenshots, it generates pre- and post-conditions specifying the action's prerequisites and intended local effect, then assesses execution success and task validity through a structured rationale and binary verdict. Training combines supervised fine-tuning, teacher-student distillation, and reward- and preference-based refinement. On 1,082 manually verified AndroidWorld transitions, Q-Reflector achieves 82.16% overall accuracy, exceeding all evaluated baselines and showing more balanced performance than GPT-5.2 and GUI-Owl-1.5-8B. On 117 MobileWorld tasks, Q-Reflector achieves task success rates of 46.15% with GUI-Owl-1.5-8B-Instruct and 35.90% with MAI-UI-8B, improving over no external reflection by 17.09 and 14.53 percentage points, respectively. Under the same condition-generation setting, it outperforms the strongest reflection baselines, GPT-5.2 and Claude Haiku 4.5, by 3.42 and 5.98 percentage points, respectively.
| Comments: | 20 pages, 6 figures. Revised and expanded version of StepReflect. An earlier version was accepted to the 1st Workshop on Multimodal Digital Agents (MMDA) at ECCV 2026 (non-archival). Code: this https URL ; Demo: this https URL |
| Subjects: | Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2608.05587 [cs.AI] |
| (or arXiv:2608.05587v2 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2608.05587 arXiv-issued DOI via DataCite |
Submission history
From: Linqiang Guo [view email]
[v1]
Thu, 6 Aug 2026 04:17:27 UTC (735 KB)
[v2]
Thu, 15 Oct 2026 07:07:07 UTC (2,412 KB)
来源：arXiv cs.AI · arxiv.org