标题： 代理工具选择中幻觉的内部表示作为指示器 显示英文标题

标题： Internal Representations as Indicators of Hallucinations in Agent Tool Selection

摘要： 大型语言模型（LLMs）在工具调用和工具使用方面表现出显著的能力，但存在幻觉问题，即选择错误的工具、提供格式错误的参数，并通过执行模拟和生成输出而不是调用专用工具或外部系统来表现出“工具绕过”行为。 这会降低基于LLM的代理在生产系统中的可靠性，因为它会导致结果不一致，并绕过安全和审计控制。 代理工具选择中的此类幻觉需要早期检测和错误处理。 与现有需要多次前向传递或外部验证的幻觉检测方法不同，我们提出了一种计算高效的框架，通过利用生成过程中使用的同一前向传递中的LLM内部表示，实现实时工具调用幻觉检测。 我们在多个领域的推理任务上评估了这种方法，证明了强大的检测性能（最高达86.4%的准确率），同时保持实时推理能力，计算开销最小，尤其在检测参数级幻觉和不适当的工具选择方面表现突出，这对可靠的代理部署至关重要。 显示英文摘要

摘要： Large Language Models (LLMs) have shown remarkable capabilities in tool calling and tool usage, but suffer from hallucinations where they choose incorrect tools, provide malformed parameters and exhibit 'tool bypass' behavior by performing simulations and generating outputs instead of invoking specialized tools or external systems. This undermines the reliability of LLM based agents in production systems as it leads to inconsistent results, and bypasses security and audit controls. Such hallucinations in agent tool selection require early detection and error handling. Unlike existing hallucination detection methods that require multiple forward passes or external validation, we present a computationally efficient framework that detects tool-calling hallucinations in real-time by leveraging LLMs' internal representations during the same forward pass used for generation. We evaluate this approach on reasoning tasks across multiple domains, demonstrating strong detection performance (up to 86.4\% accuracy) while maintaining real-time inference capabilities with minimal computational overhead, particularly excelling at detecting parameter-level hallucinations and inappropriate tool selections, critical for reliable agent deployment.