混合检索方法提升领域问答准确率，适合企业级知识系统。
Domain-specific Question Answering with Hybrid Search
- 用微调稠密检索+关键词稀疏搜索混合方案
- 融合余弦相似度、BM25和域名匹配得分，可调权重
- 在企业场景中兼顾精度与上下文可靠性
领域特定问答是不断发展的领域，需要专门解决方案应对独特挑战。本文表明，将微调的稠密检索器与基于关键词的稀疏搜索方法相结合，能显著提升性能。系统采用线性组合方式融合相关性信号，包括稠密检索的余弦相似度、BM25分数以及域名匹配，各信号具有可调增强参数。实验结果表明，该混合方法优于单一检索系统，在保持强上下文关联性的同时提升了准确率。这些发现表明，通过加权评分整合多种检索方法，能有效应对企业环境中领域特定问答的复杂性。
原文摘要 · Abstract (English)
Domain specific question answering is an evolving field that requires specialized solutions to address unique challenges. In this paper, we show that a hybrid approach combining a fine-tuned dense retriever with keyword based sparse search methods significantly enhances performance. Our system leverages a linear combination of relevance signals, including cosine similarity from dense retrieval, BM25 scores, and URL host matching, each with tunable boost parameters. Experimental results indicate that this hybrid method outperforms our single-retriever system, achieving improved accuracy while maintaining robust contextual grounding. These findings suggest that integrating multiple retrieval methodologies with weighted scoring effectively addresses the complexities of domain specific question answering in enterprise settings.
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。