真相还是海市蜃楼？基于LLM-Oasis迈向端到端事实性评估
Truth or Mirage? Towards End-to-End Factuality Evaluation with LLM-Oasis
AI总结：
本文针对LLM事实性评估资源规模小、任务单一等局限，构建了目前最大的端到端事实性评估数据集LLM-Oasis，实验表明GPT-4o在该任务中准确率仅达60%，为未来研究提供了新基准。
摘要：
自大型语言模型（LLMs）引入以来，自然语言生成（NLG）任务的性能得到了显著提升，包括文本摘要和机器翻译。然而，LLMs 仍会产生包含幻觉的输出，即未基于事实信息的内容。因此，开发评估 LLMs 事实性的方法变得迫在眉睫。\n实际上，近期已出现了一些用于事实性评估的资源。尽管具有挑战性，但这些资源面临以下一个或多个局限性：(i) 它们针对特定任务或领域定制；(ii) 规模有限，从而阻碍了新的事实性评估器的训练；(iii) 它们为更简单的验证任务而设计，例如声明验证。\n为了解决这些问题，我们引入了 LLM-Oasis，据我们所知，这是用于训练端到端事实性评估器的最大资源。LLM-Oasis 的构建过程为：从 Wikipedia 提取声明，篡改其中部分声明，并生成事实性与非事实性文本对。随后，我们依赖人工标注员来验证数据集的质量，并创建一个黄金标准测试集，用于对事实性评估系统进行基准测试。\n我们的实验表明，LLM-Oasis 对最先进的 LLMs 构成了重大挑战，其中 GPT-4o 在我们提出的端到端事实性评估任务中的准确率最高达到 60%，凸显了其推动该领域未来研究的潜力。
英文摘要：
After the introduction of Large Language Models (LLMs), there have been substantial improvements in the performance of Natural Language Generation (NLG) tasks, including Text Summarization and Machine Translation. However, LLMs still produce outputs containing hallucinations, that is, content not grounded in factual information. Therefore, developing methods to assess the factuality of LLMs has become urgent. Indeed, resources for factuality evaluation have recently emerged. Although challenging, these resources face one or more of the following limitations: (i) they are tailored to a specific task or domain; (ii) they are limited in size, thereby preventing the training of new factuality evaluators; (iii) they are designed for simpler verification tasks, such as claim verification. To address these issues, we introduce LLM-Oasis, to the best of our knowledge the largest resource for training end-to-end factuality evaluators. LLM-Oasis is constructed by extracting claims from Wikipedia, falsifying a subset of these claims, and generating pairs of factual and unfactual texts. We then rely on human annotators to both validate the quality of our dataset and to create a gold standard test set for benchmarking factuality evaluation systems. Our experiments demonstrate that LLM-Oasis presents a significant challenge for state-of-the-art LLMs, with GPT-4o achieving up to 60% accuracy in our proposed end-to-end factuality evaluation task, highlighting its potential to drive future research in the field.
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。
