Beyond Atomic Tokens: Factorizing Syllables for Language Model Pretraining
本文提出了一种基于音素的分词器，通过将音节分解为声母、韵母和声调三个部分来解决传统分词方法忽视音节内部结构的问题，从而在多种语言理解任务中取得了与现有模型相当或更好的效果。
0 citationsRead paper
Selected work
Representative Papers
ExpertHTR: Unified Handwritten Text Recognition with Multi-Task Learning and Sparse Mixture-of-Experts
为解决手写文本识别资源分散、标注格式不一的问题，提出ExpertHTR框架，通过多任务学习和稀疏混合专家模型提升识别效果。
0 citationsRead paper
A Verifier-Guided Explainable Reasoning Framework with Gold-Anchored QLoRA, Task-Aware Mixture-of-Experts, and Group-Relative RLVR
本文提出一种验证器引导的可解释推理框架，通过黄金锚定QLoRA、任务感知混合专家系统和组相对RLVR方法，提高大语言模型在教育问答中的解释性和准确性。
0 citationsRead paper
Asymmetric Paired-Annotation Learning for Multi-Structure ULF Pediatric Brain MRI Segmentation
为解决0.064T下儿科脑MRI分割难题，提出AURA方法，利用非对称监督策略处理高场和低场标注差异。
0 citationsRead paper
CG-GLORE: A Conjugate Gradient-Based Global-Local Regularization Network for Sparse-View CT Reconstruction
This study addresses the ill-posed inverse problem in sparse-view CT reconstruction and the poor convergence of existing deep learning methods by proposing a compact deep unfolding framework inspired by second-order optimization. By constructing a structured Hessian proxy with a conjugate gradient solver and designing a global-local regularization module that integrates convolutional features with Nyström attention, the method effectively models image priors. Experiments on AAPM and DeepLesion datasets demonstrate stable convergence, significant noise power reduction, and enhanced visual fidelity. Achieving superior quantitative metrics compared to state-of-the-art approaches, this work provides an efficient and reliable solution for sparse-view CT reconstruction.
0 citationsRead paper
Recent publications
Latest Papers
Beyond Atomic Tokens: Factorizing Syllables for Language Model Pretraining
本文提出了一种基于音素的分词器，通过将音节分解为声母、韵母和声调三个部分来解决传统分词方法忽视音节内部结构的问题，从而在多种语言理解任务中取得了与现有模型相当或更好的效果。
0 citationsRead paper
ExpertHTR: Unified Handwritten Text Recognition with Multi-Task Learning and Sparse Mixture-of-Experts
为解决手写文本识别资源分散、标注格式不一的问题，提出ExpertHTR框架，通过多任务学习和稀疏混合专家模型提升识别效果。
0 citationsRead paper
A Verifier-Guided Explainable Reasoning Framework with Gold-Anchored QLoRA, Task-Aware Mixture-of-Experts, and Group-Relative RLVR
本文提出一种验证器引导的可解释推理框架，通过黄金锚定QLoRA、任务感知混合专家系统和组相对RLVR方法，提高大语言模型在教育问答中的解释性和准确性。
0 citationsRead paper
Asymmetric Paired-Annotation Learning for Multi-Structure ULF Pediatric Brain MRI Segmentation
为解决0.064T下儿科脑MRI分割难题，提出AURA方法，利用非对称监督策略处理高场和低场标注差异。
0 citationsRead paper
CG-GLORE: A Conjugate Gradient-Based Global-Local Regularization Network for Sparse-View CT Reconstruction
This study addresses the ill-posed inverse problem in sparse-view CT reconstruction and the poor convergence of existing deep learning methods by proposing a compact deep unfolding framework inspired by second-order optimization. By constructing a structured Hessian proxy with a conjugate gradient solver and designing a global-local regularization module that integrates convolutional features with Nyström attention, the method effectively models image priors. Experiments on AAPM and DeepLesion datasets demonstrate stable convergence, significant noise power reduction, and enhanced visual fidelity. Achieving superior quantitative metrics compared to state-of-the-art approaches, this work provides an efficient and reliable solution for sparse-view CT reconstruction.
0 citationsRead paper