arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2604.01535cs.CL
收藏

多读多想:重新审视网络代理中的观察缩减

Read More, Think More: Revisiting Observation Reduction for Web Agents

NEC Corporation(日本电气股份有限公司)

Masafumi Enomoto, Ryoma Obara, Haochen Zhang, Masafumi Oyamada

AI总结:

本文研究网络代理中观察缩减的优化问题,发现模型能力和思考token预算影响观察表示选择,高能力模型适合详细HTML观察,低能力模型适合紧凑的访问树观察,同时引入diff-based表示提升效率。

摘要:

基于大型语言模型(LLMs)的网络代理依赖网页观察(通常表示为HTML)来识别可用动作和规划后续步骤。先前工作将HTML的冗长视为性能障碍,并采用观察缩减作为标准做法。我们重新审视这一趋势,证明最优的观察表示取决于模型能力和思考token预算:(1)紧凑的观察(访问树)更适合低能力模型,而详细的观察(HTML)对高能力模型更有利,增加思考token进一步放大HTML的优势。(2)我们的错误分析表明,高能力模型利用HTML中的布局信息实现更好的动作定位,而低能力模型在更长输入下会增加幻觉。我们还发现,在大多数模型和设置中,结合观察历史可提升性能,而diff-based表示提供了一种token高效的替代方案。基于这些发现,我们提出实践指南:根据模型能力和思考token预算适配观察表示,并使用diff-based表示结合观察历史。

英文摘要:

Web agents based on large language models (LLMs) rely on observations of web pages -- commonly represented as HTML -- as the basis for identifying available actions and planning subsequent steps. Prior work has treated the verbosity of HTML as an obstacle to performance and adopted observation reduction as a standard practice. We revisit this trend and demonstrate that the optimal observation representation depends on model capability and thinking token budget: (1) compact observations (accessibility trees) are preferable for lower-capability models, while detailed observations (HTML) are advantageous for higher-capability models; moreover, increasing thinking tokens further amplifies the benefit of HTML. (2) Our error analysis suggests that higher-capability models exploit layout information in HTML for better action grounding, while lower-capability models suffer from increased hallucination under longer inputs. We also find that incorporating observation history improves performance across most models and settings, and a diff-based representation offers a token-efficient alternative. Based on these findings, we suggest practical guidelines: adaptively select observation representations based on model capability and thinking token budget, and incorporate observation history using diff-based representations.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑