Skip to content
TheChatArchiveThe Chat Archive
gpt-4o-2024-08-06· OpenAIAi2 WildChat-4.8M (AI2)

You are a precise translator from English to Chinese, specializing in plasma…

  1. User

    System: You are a precise translator from English to Chinese, specializing in plasma physics texts. Translate the given text accurately without adding any explanations or alternatives. Keep person names unchanged. Provide your response in JSON format with a key 'translated_text'. User: Translate this text to Chinese: However, current tool-augmented LLMs primarily focus on simple function-calling scenarios with limited diversity and complexity [16]. Table 1 provides an overview of the data statistics utilized in these representative tool-augmented LLMs. Although satisfying performance has been achieved on the single function calling that executes one API in one turn, the capabilities of parallel and dependent function calls are largely overlooked. Furthermore, the constrained API domains, simplistic parameter types, and uniform data formats may hinder the applicability of the function calling to more complex, real-world tasks. Executing real-world function calls requires precise API selection and accurate parameter configuration, both of which are closely tied to the accuracy of the underlying data. Yet ensuring data accuracy for tooling remains a challenging problem, particularly when dealing with diverse and complex tasks. Assistant:

  2. Assistant
    {
        "translated_text": "然而,当前工具增强的LLM主要集中于具有有限多样性和复杂性的简单函数调用场景[16]。表1概述了这些代表性工具增强LLM中使用的数据统计。虽然在单个函数调用中取得了令人满意的性能,其在一次轮次中执行一个API,但并行和依赖函数调用的能力在很大程度上被忽视。此外,受限的API领域、简单的参数类型和统一的数据格式可能会阻碍函数调用在更复杂、实际任务中的适用性。执行实际的函数调用需要精确的API选择和准确的参数配置,这两者都与基础数据的准确性密切相关。然而,确保工具使用的数据准确性仍然是一个具有挑战性的问题,特别是在处理多样化和复杂的任务时。"
    }