Avenir-Web: 人类经验模拟的多模态网络代理与接地专家混合
Avenir-Web: Human-Experience-Imitating Multimodal Web Agents with Mixture of Grounding Experts
AI总结：
Avenir-Web通过融合接地专家与经验模仿规划，实现了在复杂网络界面中可靠执行长周期任务的开源网络代理新纪录。
摘要：
尽管多模态大语言模型取得了进展，但自主网络代理仍难以在复杂的动态网络界面中可靠地执行长周期任务。现有代理往往面临不准确的元素接地、缺乏特定网站的流程知识以及不稳定长期任务跟踪和记忆，尤其是在处理复杂的文档对象模型结构时。为了解决这些限制，我们引入了Avenir-Web，这是一种在现实世界部署中实现Online-Mind2Web基准新开放源状态的网络代理。Avenir-Web利用接地专家混合、经验模仿规划来整合流程先验知识，并结合任务跟踪检查表和自适应记忆，以实现在多样化用户界面范式中的稳健无缝交互。我们在Online-Mind2Web上评估了Avenir-Web，这是一个严格检验实时和以用户为中心的网络任务的基准。我们的结果表明，Avenir-Web显著超越了先前的开源代理，并达到了与顶级专有模型相当的性能，从而在实时网站上建立了新的开源状态。
英文摘要：
Despite advances in multimodal large language models, autonomous web agents still struggle to reliably execute long-horizon tasks on complex and dynamic web interfaces. Existing agents often suffer from inaccurate element grounding, the absence of site-specific procedural knowledge, and unstable long-term task tracking and memory, particularly when operating over complex Document Object Model structures. To address these limitations, we introduce Avenir-Web, a web agent that achieves a new open-source state of the art on the Online-Mind2Web benchmark in real-world deployment. Avenir-Web leverages a Mixture of Grounding Experts, Experience-Imitation Planning for incorporating procedural priors, and a task-tracking checklist combined with adaptive memory to enable robust and seamless interaction across diverse user interface paradigms. We evaluate Avenir-Web on Online-Mind2Web, a rigorous benchmark of live and user-centered web tasks. Our results demonstrate that Avenir-Web significantly surpasses prior open-source agents and attains performance parity with top-tier proprietary models, thereby establishing a new open-source state of the art for reliable web agents on live websites.
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。
