用多阶段推理解决用户模糊点餐需求，推荐更准更全。
From "What to Eat?" to Perfect Recipe: ChefMind's Chain-of-Exploration for Ambiguous User Intent in Recipe Recommendation
- 通过探索链逐步澄清模糊请求，生成结构化条件
- 在真实数据上评分达8.7，显著优于单一模型方案
- 适合需要精准、详细食谱推荐的场景
个性化食谱推荐面临模糊用户意图、语义准确性不足和细节覆盖不全的挑战。本文提出ChefMind，一种融合探索链（CoE）、知识图谱（KG）、检索增强生成（RAG）与大语言模型（LLM）的混合架构。CoE将模糊查询转化为结构化条件，KG提供语义推理与可解释性，RAG补充上下文烹饪细节，LLM整合输出生成连贯推荐。在下厨房（Xiachufang）数据集及人工标注查询上评估，相比仅用LLM、KG或RAG的基线模型，ChefMind在准确率、相关性、完整性和清晰度上均表现更优，平均得分8.7，而消融模型为6.4–6.7。同时，未处理查询比例降至1.6%，展现出对模糊需求的强大鲁棒性。
原文摘要 · Abstract (English)
Personalized recipe recommendation faces challenges in handling fuzzy user intent, ensuring semantic accuracy, and providing sufficient detail coverage. We propose ChefMind, a hybrid architecture combining Chain of Exploration (CoE), Knowledge Graph (KG), Retrieval-Augmented Generation (RAG), and a Large Language Model (LLM). CoE refines ambiguous queries into structured conditions, KG offers semantic reasoning and interpretability, RAG supplements contextual culinary details, and LLM integrates outputs into coherent recommendations. We evaluate ChefMind on the Xiachufang dataset and manually annotated queries, comparing it with LLM-only, KG-only, and RAG-only baselines. Results show that ChefMind achieves superior performance in accuracy, relevance, completeness, and clarity, with an average score of 8.7 versus 6.4-6.7 for ablation models. Moreover, it reduces unprocessed queries to 1.6%, demonstrating robustness in handling fuzzy demands.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。