研究人员开发了一种半自动工作流程来评估由大型语言模型 (LLM) 生成或增强的食谱数据的合理性。该方法应用于印度食品知识图谱，识别并纠正常见的 LLM 错误，例如幻觉成分、不正确的数量和文化上不合理的组合。该流程结合了形式语法、统计启发式方法和基于 Transformer 的连贯性建模，以确保机器可读食品知识的准确性和适用性。 AI
影响 增强了 LLM 生成内容在知识图谱构建和专业领域下游应用中的可靠性。
排序理由 该集群包含一篇学术论文，详细介绍了验证 LLM 生成数据的新方法。
在 arXiv cs.IR (Information Retrieval) 阅读 →
AI 生成摘要 · Google Gemini · 来自 3 个来源。 我们如何撰写摘要 →
完整方法见我们的编辑标准。
arXiv:2608.29249v1 Announce Type: new Abstract: The online culinary ecosystem is increasingly populated by recipe content generated, modified, or summarized by Large Language Models (LLMs). While often plausible, such outputs may contain hallucinated ingredients, misrepresented q…
The online culinary ecosystem is increasingly populated by recipe content generated, modified, or summarized by Large Language Models (LLMs). While often plausible, such outputs may contain hallucinated ingredients, misrepresented quantities, or culturally implausible combination…
The online culinary ecosystem is increasingly populated by recipe content generated, modified, or summarized by Large Language Models (LLMs). While often plausible, such outputs may contain hallucinated ingredients, misrepresented quantities, or culturally implausible combination…