Evolving Agent Interfaces
[Sztuczna inteligencja] Autorzy: Saaket Agashe, Yuzhe Yang, Gonzalo Gonzalez-Pumariega i 5 więcej
Computer-use agents are limited by the interfaces humans give them. An agent may type a spreadsheet cell by cell when a single line of code would do, or print all two thousand lines of a file through a shell to find one function. Both the interfaces and the knowledge of when to use each are built by hand, which cannot keep pace with new applications, workflows, and models. We argue that interface design should itself be an agent capability. We introduce Agent S4, an autonomous computer-use scientist that builds Evolving Agent Interfaces from task experience. A meta-agent examines task agents' trajectories on synthetic training tasks, without success labels, and constructs and revises interfaces and usage knowledge in a repository. Population-based evolution then selects which repositories to develop further by their performance on validation tasks the meta-agent never sees, favoring cha…
- Kategoria arxiv
- cs.AI
- Kategoria
- Sztuczna inteligencja