Tag
该论文提出一个对比学习框架，将EEG信号与冻结的预训练文本编码器的嵌入空间对齐，用于解码想象言语，并在Chisco数据集上验证了其在固定意图解码、相似度检索和语义结构探针方面的能力。
CliQ-RRG is a two-stage framework for QA-style radiology report generation that uses disease-aware visual-textual alignment, prior-guided attention across scans, and GraphRAG-augmented LLM restructuring to produce clinically grounded reports, outperforming prior methods on two chest X-ray benchmarks.
This paper presents 3D masked autoencoders for volumetric microscopy data, demonstrating that 3D modeling outperforms 2D max-projection and slice-based variants on downstream single-cell tasks, with cross-modal alignment to a protein language model further improving performance.