HalluShift：测量大型语言模型中幻觉检测的分布偏移
HalluShift: Measuring Distribution Shifts towards Hallucination Detection in LLMs
AI总结：
针对大型语言模型（LLMs）的幻觉问题，本研究提出HalluShift方法，通过分析LLM生成响应的内部状态空间和标记概率的分布偏移检测幻觉，在多个基准数据集上性能优于现有基线。
摘要：
大型语言模型（LLMs）因其能够针对多个领域的给定提示生成创新响应而近期受到广泛关注。然而，LLMs常受幻觉固有局限影响，生成信息错误但结构良好、连贯的响应。本研究假设幻觉源于LLMs的内部动态。观察表明，LLMs在生成段落时，会在响应的细微部分偏离事实准确性，最终转向错误信息。该现象与人类认知相似：个体可能在保持逻辑连贯的同时产生幻觉，将不确定性嵌入言语的次要片段。为进一步研究，我们提出创新方法HalluShift，用于分析LLM生成响应的内部状态空间和标记概率的分布偏移。我们的方法在多个基准数据集上的性能优于现有基线。代码库可在https://github.com/sharanya-dasgupta001/hallushift获取。
英文摘要：
Large Language Models (LLMs) have recently garnered widespread attention due to their adeptness at generating innovative responses to the given prompts across a multitude of domains. However, LLMs often suffer from the inherent limitation of hallucinations and generate incorrect information while maintaining well-structured and coherent responses. In this work, we hypothesize that hallucinations stem from the internal dynamics of LLMs. Our observations indicate that, during passage generation, LLMs tend to deviate from factual accuracy in subtle parts of responses, eventually shifting toward misinformation. This phenomenon bears a resemblance to human cognition, where individuals may hallucinate while maintaining logical coherence, embedding uncertainty within minor segments of their speech. To investigate this further, we introduce an innovative approach, HalluShift, designed to analyze the distribution shifts in the internal state space and token probabilities of the LLM-generated responses. Our method attains superior performance compared to existing baselines across various benchmark datasets. Our codebase is available at https://github.com/sharanya-dasgupta001/hallushift.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
