Clinical Concept Centers in LLMs
arXiv cs.CL ·
Large language models are increasingly used in clinical settings. However, research into the reliability and performance of these models has focused almost entirely on the langu…
原文内容 · 暂无该语言译文AI News · 2026-10-05
按本站收录日期（UTC）归档，不代表事件发生于当天；原公告、聚合源发布与收录时间分别标注。
返回资讯列表CLIMB: Confidence-Guided Complementary Evidence for Multimodal Retrieval-Augmented Generation
arXiv cs.CL ·
Multimodal large language models (MLLMs) have shown strong visual reasoning abilities, but knowledge-intensive visual question answering often requires external textual evidence…
原文内容 · 暂无该语言译文On Representational Alignment among Embodied Agents
arXiv cs.RO ·
Embodied agents interacting with the same physical process may maintain heterogeneous, asynchronous, and observer-relative representations. Rather than assuming that such repres…
原文内容 · 暂无该语言译文NEEDLEWORK: Offline Rewriting of Robot Data with Verified Local Stitches
arXiv cs.RO ·
Robot demonstrations may contain useful behavior even when individual episodes are inefficient or unsuccessful. Trajectory stitching offers a way to compose these behaviors into…
原文内容 · 暂无该语言译文