Guides That Cause Actions: An Offline Study of Guide-Action Mutual Reinforcement in Multimodal Web Agents
作者: Chengguang Gan, Yunhao Liang, QingHao Zhang, Shiwen Ni
分类: cs.CL
发布日期: 2026-09-23
💡 一句话要点
提出WebMRE基准以研究多模态网络代理中的引导-动作相互强化问题
🎯 匹配领域: 支柱九:具身大模型 (Embodied Foundation Models)
关键词: 多模态网络代理 引导-动作相互强化 离线基准 智能助手 决策优化
📋 核心要点
- 现有方法在实时环境中评估网络代理时,环境状态和评判模型漂移,导致难以控制训练现象的研究。
- 论文提出WebMRE离线基准,通过将引导句与动作配对,研究它们之间的相互强化效应。
- 实验结果显示,联合解码引导句显著提升了动作选择的准确性,且模型在多个离线指标上超越了现有基线。
📝 摘要(中文)
网络代理通常在实时环境中评估,环境状态和评判模型在不同运行间漂移,使得相同检查点难以重现相同分数,导致训练现象的控制研究变得不切实际。本文提出WebMRE,这是一个离线基准,包含541个任务和5,293个步骤,源自成功的WebArena轨迹,具有完全审核的测试标签和确定性的评分协议。每一步将人类导向的引导句与具体动作配对,使得首次研究它们之间的相互强化效应成为可能。实验结果表明,联合解码引导句显著提升了动作选择的准确性,验证了引导句在动作决策中的因果作用。
🔬 方法详解
问题定义:本文旨在解决网络代理在实时环境中评估时的不可重复性问题,现有方法难以控制训练现象,导致研究结果不稳定。
核心思路:提出WebMRE离线基准,通过将引导句与具体动作配对,研究它们之间的相互强化效应,提供可重复的实验环境。
技术框架:整体架构包括数据收集、引导句与动作配对、离线评估和结果分析等模块,确保每次运行的评分一致性。
关键创新:引入引导句作为因果通道,而非仅仅作为评论,首次证明引导句在动作决策中的重要性,显著提升了动作准确性。
关键设计:采用确定性的评分协议,确保每个检查点在每次运行中获得相同分数;通过中介分析验证引导句的因果作用,设计了多种引导句的变体以测试其对动作选择的影响。
🖼️ 关键图片
📊 实验亮点
实验结果表明,联合解码引导句显著提升了动作选择的准确性,Qwen3.5-4B模型的提升幅度为0.9和0.2分,而Qwen3.5-9B模型的提升幅度为1.7和2.2分。此外,优化后的模型在所有离线指标上超越了GPT-5.5、Claude Opus 4.8和Gemini 3.5 Flash。
🎯 应用场景
该研究的潜在应用领域包括智能助手、自动化客服和多模态交互系统等。通过优化引导-动作相互强化机制,可以提升用户体验和系统的决策能力,未来可能在更广泛的人工智能应用中发挥重要作用。
📄 摘要(原文)
Web agents are usually evaluated in live environments, where environment state and judge models drift between runs, so the same checkpoint rarely reproduces the same score, making controlled studies of training phenomena impractical. We present WebMRE, an offline benchmark of 541 tasks and 5,293 steps derived from successful WebArena trajectories, with fully audited test labels and a deterministic protocol that scores a checkpoint identically on every run without any environment. Each step pairs a human oriented guide sentence with a grounded action, enabling the first study of the mutual reinforcement effect between them in web agents. Averaged over three seeds the effect holds for both models in both decoding orders and grows with scale: jointly decoding a guide lifts element selection over an action only reference by 0.9 and 0.2 points for Qwen3.5-4B and by 1.7 and 2.2 points for Qwen3.5-9B. A mediation analysis shows that the guide is a causal channel rather than commentary: forcing the gold guide as a decoding prefix lifts action accuracy from .422 to .684, another step's guide collapses it to .055, and a paraphrase that renames the target still recovers half of the gain, so the channel carries instruction meaning and not only the label string. The same channel yields an offline reward that only a replayable protocol makes computable, though optimizing it from a strong checkpoint brings no gain yet. Our fine tuned models outperform GPT-5.5, Claude Opus 4.8, and Gemini 3.5 Flash, run zero shot, on every offline metric.