Real-time hallucination detection and intervention in medical LLMs via calibrated hidden-state probes [0.03%]
通过校准隐藏状态探针实时检测和干预医学LLM中的幻觉现象
Zhengqiu Yu,Xiangrong Liu
Zhengqiu Yu
Objective: Most hallucination mitigation for large language models (LLMs) operates post-hoc, leaving safety-critical clinical deployment without real-time warning capability. We present a calibrated hidden-state probing p...
A multi-modal temporal fusion transformer for comprehensive decision support across the emergency care trajectory [0.03%]
一种多模式时间融合变压器，可沿急救护理流程提供全面的决策支持
Dong Hyun Choi,Ki Jeong Hong,Hyun Wook Ryoo et al.
Dong Hyun Choi et al.
Introduction: Emergency care in the emergency department (ED) requires continuous, multifaceted decision-making based on evolving clinical information. This study aimed to develop and externally validate a comprehensive E...
A scoping review of algorithmic equity, data diversity, and inclusive design in the transformer era of clinical NLP [0.03%]
转换器时代临床自然语言处理中的算法公平性、数据多样性及包容性设计的综述
Sheri Grach,Abeer Badawi,Farah Ahmad et al.
Sheri Grach et al.
Background: The rapid digitization of healthcare has positioned transformer-based natural language processing (NLP) models as powerful tools for managing clinical textual data. However, their integration into practice rai...
Multimodal AI in healthcare: Review of vision-language foundation models for real-world medical applications [0.03%]
多模态人工智能在医疗领域的应用：面向实际医学应用的视觉-语言基础模型综述
Taha Razzaq,Murtaza Taj,Asim Iqbal
Taha Razzaq
The emergence of foundation models has marked a transformative shift in AI, enabling robust generalization across diverse downstream tasks through putative zero-shot learning. Large Language Models and Vision-Language Models have demonstrat...
scCLIP: A contrastive masked-reconstruction framework for paired single-cell multi-omics integration [0.03%]
scCLIP：一种配对单细胞多组学整合的对比掩码重建框架
Xiang Xu,Li Du
Xiang Xu
Paired biomedical assays increasingly measure different molecular or clinical views from the same sample. The statistical problem is simple to state but hard to solve: the views often have different dimensions, noise models, and dynamic ran...
Ihor Kendiukhov
Ihor Kendiukhov
Objective: Single-cell foundation models such as scGPT have been promoted as representations of gene regulation, but their advertised regulatory signal has been judged largely from attention weights, which are correlation...
CoAff-DTI: Fine-grained drug-target interaction prediction using pre-trained language models and affinity-guided mechanisms [0.03%]
基于预训练语言模型和亲和力引导机制的细粒度药物-目标相互作用预测
Jia Peng,Xiaoyu Liu,Lei Wang et al.
Jia Peng et al.
Accurate prediction of drug-target interactions (DTI) is essential for drug discovery. Despite the success of pre-trained language models (PLMs) in learning robust molecular and protein representations, a fundamental challenge remains in ch...
Evaluation of temporal preservation in synthetic longitudinal patient data [0.03%]
合成纵向患者数据中时间信息的评估方法研究
Katariina Perkonoja,Parisa Movahedi,Antti Airola et al.
Katariina Perkonoja et al.
Objective: This study introduces a set of metrics for evaluating temporal preservation in synthetic longitudinal patient data, defined as artificially generated data that mimic real patients' repeated measurements over ti...
ARKE: An ontology-driven framework for automated mapping of local radiology procedure terms to the LOINC-RadLex playbook using large language model [0.03%]
基于本体的框架自动映射本地放射学程序术语到LOINC-RadLex剧本的框架（ARKE）
Sumin Lee,Kyulee Jeon,Joon Seok Lim et al.
Sumin Lee et al.
Objective: To develop an ontology-driven framework that standardizes heterogeneous local radiology procedure names by decomposing them into semantic components and aligning them to LOINC/RSNA Radiology Playbook codes usin...
A validation-driven training controller for cross-lingual biomedical NER via reinforcement learning-based adaptive loss weighting [0.03%]
基于强化学习的自适应损失赋权的跨语言生物医学命名实体识别验证驱动训练控制器
Chensen Zhang,Wei Liao
Chensen Zhang
Objective: Cross-lingual biomedical named entity recognition (BioNER) remains challenging in low-resource settings due to scarce annotations, heterogeneous corpora, and severe label imbalance dominated by the O class. We ...