Getting the Parameters Right: A Difficulty-Graded Benchmark and Probe-Guided Training for LLM Tool Calls 文章
详细信息
- 来源站点
- ArXiv CS.AI
- 作者
- Guoyao Yu, Xiaoqing Sun, Ziqi Huang, Shaojing Fan, Zhongyi Zhang, Xiaomeng Hu, Xiaobo Xue, Yangyang Shi, Xiong Xiao, Yang Song, Biao Lyu, Rong Wen, Xing Li, Qinming He, Shunming Zhu, Zhenguang Liu
- 文章类型
- PAPER
- 语言
- en
- 发布日期
- 2026-08-05
摘要
arXiv:2608.03071v1 Announce Type: new Abstract: Large language model agents derive much of their capability from tool use. Existing research on tool use has largely focused on selecting the right tool and orchestrating the order of calls. However, correctly filling the parameters of a tool call is equally critical for successful execution and has received far less attention. In domains such as cloud networking, even frontier models correctly complete fewer than half of tool calls. Inspired by recent analyses showing that LLM hidden states encode rich information about model predictions, we discover that while the model generates a parameter value, its hidden state contains a strong correctness signal: a simple linear probe can accurately predict whether the value will be correct.