阿斯塔纳时间
AI QAZAQSTANKZ · AI测试版

哈萨克斯坦人工智能与高科技领域的智能门户

Research1 分钟阅读

内容来自原始来源：

arXiv — cs.AI

在 arXiv — cs.AI 阅读全文

本内容来自公开来源。若发布内容影响您的权利 → 申请删除信息

标题为英语Choosing Before Acting: Comparative Value Estimation for Long-Horizon Tool-Use Agents

Large language models (LLMs) rely on long-horizon tool invocation sequences for complex tasks, where each invocation can alter the task state and condition subsequent decisions. In long-horizon tool use, final-outcome rewards provide weak credit assignment over long interaction…

arXiv — cs.AI作者: Yu Li, Zheng Zhang, Xin Liu, Shengtian Yang, Guangfeng Cai, Lei Feng2026年10月5日 (10月5日)
提供 1 种语言版本

AI 摘要：要点

  • Large language models (LLMs) rely on long-horizon tool invocation sequences for complex tasks, where each invocation can alter the task state and condition subsequent decisions
  • In long-horizon tool use, final-outcome rewards provide weak credit assignment over long interaction…
Research
KZ·AI 4F48
Инсайт дня

Choosing Before Acting: Comparative Value Estimation for Long-Horizon Tool-Use Agents

标签：#arXiv
Поделиться:

相关文章

返回列表