新框架允许冻结的 AI 模型从部署经验中学习

作者 PulseAugur 编辑部 · [1 个来源] · 2026-10-19 04:00

研究人员开发了一个新颖的框架，使冻结的大型语言和视觉语言模型能够在无需访问模型权重的情况下从部署经验中学习。这种模型无关的方法通过一个用于推理和工具使用的 Skill、一个用于存储可靠事实的 Knowledge Memory 以及一个用于视觉示例的 Multimodal Knowledge Base 来整合外部专业知识。该系统在各种基准和基础模型上，将医疗任务的性能提高了高达 34.2%，同时还能泛化到未见过的情况并迁移到其他模型。 AI

影响使医疗等关键领域的已部署 AI 模型能够持续改进，而无需昂贵的再培训。

排序理由该集群包含一篇详细介绍新 AI 框架的学术论文。[lever_c_demoted from research: ic=1 ai=1.0]

在 arXiv cs.AI 阅读 →

AI 生成摘要 · Google Gemini · 来自 1 个来源。我们如何撰写摘要 →

本文如何被排名

Signal score: 10 / 100; Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket: Tool; 该集群包含一篇详细介绍新 AI 框架的学术论文。[lever_c_demoted from research: ic=1 ai=1.0]
Source corroboration: Single-source cluster; Only one publisher covered this so far. Single-source stories can still rank when the publisher is high-authority, but they lack cross-source corroboration.
Topics: paper, model release; Editorial topic classification. Feeds into how the story surfaces on /topic/<slug> hub pages and into the per-entity coverage mix.
AI-industry relevance: High; Clearly on-topic for AI-industry coverage.
Story freshness: Same-day; Cluster formed today. Ranking reflects the current source set at time of score.

完整方法见我们的编辑标准。

报道来源 [1]

arXiv cs.AI TIER_1 English(EN) · Yexiao He, Yucheng Tang, Pengfei Guo, Yufan He, Andriy Myronenko, Can Zhao, Ang Li, Daguang Xu, Dong Yang · 2026-10-19 04:00

模型冻结，专业知识演进：面向多模态医疗AI的部署经验模型无关学习

arXiv:2610.09146v2 Announce Type: replace Abstract: Large language models (LLMs) and vision-language models (VLMs) are usually frozen after deployment, so they do not learn from the cases they solve. This is especially concerning in medicine, where new clinical evidence, updated …

本文如何被排名

报道来源 [1]

模型冻结，专业知识演进：面向多模态医疗AI的部署经验模型无关学习

相关实体

相关话题