PulseAugur
实时 19:13:37
English(EN) Knowledge-Graph Grounding Helps LLMs Only for Out-of-Training Knowledge: A Controlled Study on Clinical Question Answering

用于医疗问答的LLM:探索新的推理提示和知识图谱接地

研究人员正在探索改进大型语言模型(LLM)在开放式医疗问答方面的能力。一种方法是使用一种名为CLINICR的思维链(CoT)推理提示,旨在模仿临床推理,并在MEDQA-OPEN等修改后的数据集上表现优于现有的5-shot CoT提示。另一项研究调查了知识图谱(KG)接地的有效性,发现它仅在所需信息超出模型训练数据范围时,特别是对于新颖或私有知识,才能显著提高LLM的准确性,而对已知事实的益处很小。 AI

影响 这些研究表明,先进的推理技术和有针对性的知识整合可以显著增强LLM在医学等专业领域的应用能力,有望在医疗保健领域带来更可靠的AI助手。

排序理由 两篇arXiv论文展示了关于改进LLM在医疗问答方面性能的新研究。

在 arXiv cs.CL 阅读 →

AI 生成摘要 · Google Gemini · 来自 3 个来源。 我们如何撰写摘要 →

用于医疗问答的LLM:探索新的推理提示和知识图谱接地

报道来源 [3]

  1. arXiv cs.CL TIER_1 English(EN) · Saeel Sandeep Nachane, Ojas Gramopadhye, Prateek Chanda, Ganesh Ramakrishnan, Kshitij Sharad Jadhav, Yatin Nandwani, Dinesh Raghu, Sachindra Joshi ·

    少样本思维链驱动推理提示大型语言模型进行开放式医学问题解答

    arXiv:2403.04890v4 Announce Type: replace Abstract: In this paper, we propose a modified version of the MedQA-USMLE dataset, named MEDQA-OPEN, which contains open-ended medical questions without options to mimic clinical scenarios, along with clinician-approved reasoned answers. …

  2. arXiv cs.CL TIER_1 English(EN) · Madhulatha Mandarapu, Sandeep Kunkunuru ·

    知识图谱增强仅对训练外知识的LLM有效:一项关于临床问答的对照研究

    arXiv:2606.22419v2 Announce Type: replace Abstract: A recent Nature Medicine study reports that general-purpose frontier LLMs outperform specialized retrieval-augmented clinical tools on medical benchmarks, and that retrieval can hurt strong models. We ask the natural follow-up: …

  3. arXiv cs.CL TIER_1 English(EN) · Sandeep Kunkunuru ·

    知识图谱增强仅对训练外知识的LLM有效:一项临床问答的对照研究

    A recent Nature Medicine study reports that general-purpose frontier LLMs outperform specialized retrieval-augmented clinical tools on medical benchmarks, and that retrieval can hurt strong models. We ask the natural follow-up: does structured knowledge-graph (KG) grounding chang…