SUGAR：利用上下文置信度实现更智能的检索
SUGAR: Leveraging Contextual Confidence for Smarter Retrieval
AI总结：
针对RAG统一检索导致的来源低效与噪声干扰问题，本文提出SUGAR，用基于上下文的熵自适应决定是否检索及单步或多步检索，从而提升问答性能和推理效率。
摘要：
考虑到大型语言模型（LLM）有限的参数化知识，检索增强生成（RAG）通过为其提供相关外部知识，已成为在一定程度上缓解幻觉问题的一种方法。然而，统一检索支持上下文会使响应生成产生来源低效问题，因为触发检索器并不总是必要的，甚至可能并不准确：当模型被嘈杂的检索内容分散注意力时，就会产生无用的答案。受这些问题启发，我们提出了语义不确定性引导的自适应检索（Semantic Uncertainty Guided Adaptive Retrieval，SUGAR），其中利用基于上下文的熵来主动决定是否进行检索，并进一步确定采用单步检索还是多步检索。实验结果表明，由语义不确定性估计引导的选择性检索提升了多种问答任务上的表现，并实现了更高效的推理。
英文摘要：
Bearing in mind the limited parametric knowledge of Large Language Models (LLMs), retrieval-augmented generation (RAG) which supplies them with the relevant external knowledge has served as an approach to mitigate the issue of hallucinations to a certain extent. However, uniformly retrieving supporting context makes response generation source-inefficient, as triggering the retriever is not always necessary, or even inaccurate, when a model gets distracted by noisy retrieved content and produces an unhelpful answer. Motivated by these issues, we introduce Semantic Uncertainty Guided Adaptive Retrieval (SUGAR), where we leverage context-based entropy to actively decide whether to retrieve and to further determine between single-step and multi-step retrieval. Our empirical results show that selective retrieval guided by semantic uncertainty estimation improves the performance across diverse question answering tasks, as well as achieves a more efficient inference.
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。
