Gespeichert in:
| Hauptverfasser: | , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2025
|
| Schlagworte: | |
| Online-Zugang: | https://arxiv.org/abs/2505.10887 |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866910181759123456 |
|---|---|
| author | Lei, Bin Kang, Weitai Zhang, Zijian Chen, Winson Xie, Xi Zuo, Shan Xie, Mimi Payani, Ali Hong, Mingyi Yan, Yan Ding, Caiwen |
| author_facet | Lei, Bin Kang, Weitai Zhang, Zijian Chen, Winson Xie, Xi Zuo, Shan Xie, Mimi Payani, Ali Hong, Mingyi Yan, Yan Ding, Caiwen |
| contents | This paper introduces \textsc{InfantAgent-Next}, a generalist agent capable of interacting with computers in a multimodal manner, encompassing text, images, audio, and video. Unlike existing approaches that either build intricate workflows around a single large model or only provide workflow modularity, our agent integrates tool-based and pure vision agents within a highly modular architecture, enabling different models to collaboratively solve decoupled tasks in a step-by-step manner. Our generality is demonstrated by our ability to evaluate not only pure vision-based real-world benchmarks (i.e., OSWorld), but also more general or tool-intensive benchmarks (e.g., GAIA and SWE-Bench). Specifically, we achieve $\mathbf{7.27\%}$ accuracy on OSWorld, higher than Claude-Computer-Use. Codes and evaluation scripts are open-sourced at https://github.com/bin123apple/InfantAgent. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2505_10887 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | InfantAgent-Next: A Multimodal Generalist Agent for Automated Computer Interaction Lei, Bin Kang, Weitai Zhang, Zijian Chen, Winson Xie, Xi Zuo, Shan Xie, Mimi Payani, Ali Hong, Mingyi Yan, Yan Ding, Caiwen Artificial Intelligence This paper introduces \textsc{InfantAgent-Next}, a generalist agent capable of interacting with computers in a multimodal manner, encompassing text, images, audio, and video. Unlike existing approaches that either build intricate workflows around a single large model or only provide workflow modularity, our agent integrates tool-based and pure vision agents within a highly modular architecture, enabling different models to collaboratively solve decoupled tasks in a step-by-step manner. Our generality is demonstrated by our ability to evaluate not only pure vision-based real-world benchmarks (i.e., OSWorld), but also more general or tool-intensive benchmarks (e.g., GAIA and SWE-Bench). Specifically, we achieve $\mathbf{7.27\%}$ accuracy on OSWorld, higher than Claude-Computer-Use. Codes and evaluation scripts are open-sourced at https://github.com/bin123apple/InfantAgent. |
| title | InfantAgent-Next: A Multimodal Generalist Agent for Automated Computer Interaction |
| topic | Artificial Intelligence |
| url | https://arxiv.org/abs/2505.10887 |