Avenir-Web: Human-Experience-Imitating Multimodal Web Agents with Mixture of Grounding Experts
This addresses the challenge of reliable web automation for users and developers, representing a strong incremental advance in open-source web agents.
The paper tackled the problem of autonomous web agents struggling with long-horizon tasks on complex web interfaces by introducing Avenir-Web, which achieved a new open-source state of the art on the Online-Mind2Web benchmark, matching top-tier proprietary models.
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.