生成AI悖论：它能解决什么，它可能无法评估
The Generative AI Paradox on Evaluation: What It Can Solve, It May Not Evaluate
AI总结：
本文研究了生成AI在评估任务中的局限性，发现LLMs在生成任务中表现优异，但在评估任务中却存在不忠实的问题，揭示了生成能力与评估能力之间的悖论。
摘要：
本文探讨了大型语言模型（LLMs）在生成任务中擅长的假设，即它们在评估任务中同样熟练。我们使用TriviaQA（Joshi等，2017）数据集评估了三个LLMs和一个开源LM在问答（QA）和评估任务中的性能。结果显示，LLMs在评估任务中的表现显著低于生成任务。有趣的是，我们发现了一些不忠实的评估实例，即模型在缺乏能力的领域准确评估答案，这凸显了需要检查LLMs作为评估者的忠实性和可信度。本研究有助于理解“生成AI悖论”（West等，2023），强调了探索生成卓越与评估能力之间的相关性，以及在模型评估中审视忠实性方面的必要性。
英文摘要：
This paper explores the assumption that Large Language Models (LLMs) skilled in generation tasks are equally adept as evaluators. We assess the performance of three LLMs and one open-source LM in Question-Answering (QA) and evaluation tasks using the TriviaQA (Joshi et al., 2017) dataset. Results indicate a significant disparity, with LLMs exhibiting lower performance in evaluation tasks compared to generation tasks. Intriguingly, we discover instances of unfaithful evaluation where models accurately evaluate answers in areas where they lack competence, underscoring the need to examine the faithfulness and trustworthiness of LLMs as evaluators. This study contributes to the understanding of "the Generative AI Paradox" (West et al., 2023), highlighting a need to explore the correlation between generative excellence and evaluation proficiency, and the necessity to scrutinize the faithfulness aspect in model evaluations.
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。
