S1-NexusAgent: a Self-Evolving Agent Framework for Multidisciplinary Scientific Research
S1-NexusAgent Team, Institute of Automation, Chinese Academy of Sciences · 2026
Read S1-NexusAgent: a Self-Evolving Agent Framework for Multidisciplinary Scientific Research by S1-NexusAgent Team, Institute of Automation, Chinese Academy of Sciences (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2602-01550.txt.
Description
The S1-NexusAgent Team proposes S1-NexusAgent, a self-evolving framework for multidisciplinary research. Current models and tool agents falter at long-horizon planning, goal upkeep, and learning from execution. It splits global planning from subtask tool execution via a hierarchical plan-and-code-act dual-loop design. It supports the model context protocol, orchestrating thousands of tools via intention-aware retrieval and hot-plug links. Object-reference sparse context isolates subtask contexts and compresses results. A critic distills strong trajectories into reusable skills, closing the self-evolution loop.
Catalog record
| CREATOR | S1-NexusAgent Team, Institute of Automation, Chinese Academy of Sciences |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 71824 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2602.01550