Saaket Agashe, Yuzhe Yang, Gonzalo Gonzalez-Pumariega, Bhuvan Chandra Koduru, Chih-Lun Li, Jiachen Yang, Ang Li, Xin Eric Wang · 2026 · arXiv (Cornell University)
DOI: 10.48550/arxiv.2610.28601 · cited by 0 works (OpenAlex) · open-access PDF
कंप्यूटर-उपयोग एजेंटों को मनुष्यों द्वारा दिए गए इंटरफ़ेसों द्वारा सीमित किया जाता है। एक एजेंट एक टेप शीट से सेल टाइप कर सकता है जब एक एकल लाइन कोड करना होगा, या एक शेल्फ के माध्यम से एक ही कार्य खोजने के लिए एक ही फ़ाइल के सभी दो हजार लाइनों को प्रिंट कर सकता है। दोनों इंटरफ़ेस और प्रत्येक का उपयोग करने के लिए ज्ञान हाथ से बनाया जाता है, जो नई अनुप्रयोगों, कार्य प्रवाहों और मॉडल के साथ गति नहीं रख सकता है। हम तर्क देते हैं कि इंटरफ़ेस डिजाइन स्वयं एक एजेंट क्षमता होनी चाहिए। हम एजेंट एस4 को पेश करते हैं, एक स्वतंत्र कंप्यूटर उपयोग वैज्ञानिक जो कार्य अनुभव से विकसित एजेंट इंटरफ़ेस का विकास करता है।
Machine translation (Meta M2M100); the English original follows.
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 changes that generalize beyond the seen tasks. On OSWorld V2, Agent S4, with Qwen3.8-27B and the evolved repository of interfaces, sets a new state of the art in open-source computer-use agents, reaching 21.3% binary score, 3.8 times the same model with fixed interfaces. This is above both GPT-5.5 (13.0%) and Claude Opus 4.8 (20.6%), at about 4.3% and 1.5% of their respective costs. On ScienceBoard, the evolved interfaces improve the success rate from 41.4% to 58.6% with the same model. These results suggest that useful interfaces can be evolved from experience, making autonomous interface design a source of improvement complementary to advances in models and harnesses. We open-source our code at https://github.com/simular-ai/agent-s.
Read in: English · Spanish · Chinese · Hindi · Arabic · Portuguese · French · German · Japanese · Russian · Korean · Indonesian · Vietnamese
Topics: Computer science, Task (project management), Interface (matter), Code (set theory), Human–computer interaction, Autonomous agent
GET https://store.agentexchange.work/gw/papers/verify?title=...&year=... (0.003 USD), fetch this record as JSON with /gw/papers/paper?id=2610.28601, or list what it cites with /gw/papers/references?id=.... Pay per call in USDC (x402) or with a 19 USD card key for 1,000 calls. Catalogue · llms.txtSources: OpenAlex (CC0), arXiv, Crossref. Agent Exchange