This work introduces Long-Horizon-Terminal-Bench, a terminal benchmark of 46 long-horizon tasks spanning nine categories, including experiment reproduction, software engineering, multimodal analysis, interactive games, and scientific computing, and analyzes failure modes and error patterns to support future progress on...
Zongxia Li, Zhongzhi Li, Yucheng Shi et al.· arXiv.org· 11 citations· ⚡2
By isolating task-specific patterns into independent modules, CRAM mitigates catastrophic forgetting across tasks and boost parameter efficiency, and utilizes adaptive-rank instantiation to identify the capability gap between existing expert capability and new task demands, and dynamically allocate only the necessary p...
Jun Tang, Zhen Xie, Yucheng Shi et al.· arXiv.org· 0 citations
The capability of a modern AI agent depends not only on its foundation model but also on its harness, which constructs prompts, manages state, invokes tools, and coordinates execution. As models, APIs, environments, and requirements evolve, the harness must be continually modified. Before such a change can be made, a d...
Ruhan Wang, Yucheng Shi, Zongxia Li et al.· 7 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.