Author name cluster

Yonghong He

Possible papers associated with this exact author name in Arrow. This page groups case-insensitive exact name matches and is not a full identity disambiguation profile.

5 papers

1 author row

JBHI Journal 2026 Journal Article

DiagR1: A Vision-Language Model Trained via Reinforcement Learning for Digestive Pathology Diagnosis

Minxi Ouyang
Lianghui Zhu
Yaqing Bao
Qiang Huang
Jingli Ouyang
Tian Guan
Xitong Ling
Jiawen Li

Multimodal large models have shown great potential in automating pathology image analysis. However, current multimodal models for gastrointestinal pathology are constrained by both data quality and reasoning transparency: pervasive noise and incomplete annotations in public datasets predispose vision-language models to factual hallucinations when generating diagnostic text, while the absence of explicit intermediate reasoning chains renders the outputs difficult to audit and thus less trustworthy in clinical practice. To address these issues, we construct a large-scale gastrointestinal pathology dataset containing both microscopic descriptions and diagnostic conclusions, and propose a prompt augmentation strategy that incorporates lesion classification and anatomical site information. This design guides the model to better capture image-specific features and maintain semantic consistency in generation. Furthermore, we employ a post-training pipeline that combines supervised fine-tuning with Group Relative Policy Optimization (GRPO) to improve reasoning quality and output structure. Experimental results on real-world pathology report generation tasks demonstrate that our approach significantly outperforms state-of-the-art open-source and proprietary baselines in terms of generation quality, structural completeness, and clinical relevance. Our solution outperforms state-of-the-art models with 18. 7% higher clinical relevance, 32. 4% improved structural completeness, and 41. 2% fewer diagnostic errors, demonstrating superior accuracy and clinical utility compared to existing solutions.

Details DOI

NeurIPS Conference 2025 Conference Paper

A Simple Linear Patch Revives Layer-Pruned Large Language Models

Xinrui Chen
Haoli Bai
Tao Yuan
ruikang liu
Kang Zhao
Xianzhi Yu
Lu Hou
Tian Guan

Layer pruning has emerged as a widely used technique for compressing large language models (LLMs). However, existing layer pruning approaches often incur substantial performance degradation. We identify the majority of this degradation to a single yet previously overlooked issue: \textit{the mismatch of activation magnitudes at the pruning interface}. The pre-interface activations exhibit significantly different scales from the post-interface ones, causing the distributional shift as it propagates through the remaining layers. To address this issue, we introduce \textsc{LinearPatch}, a lightweight and plug-and-play technique that fuses two operations into one matrix multiply at the pruning interface: (i) a Hadamard transformation that suppresses massive outliers at particular tokens and (ii) a channel-wise scaling that aligns activation statistics. On LLaMA-3-8B, \textsc{LinearPatch} preserves up to \textbf{94. 15\%} of the original model's performance when pruning 5 out of 32 layers, outperforming the previous state of the art by \textbf{4\%}. The patch can be further refined with 5K unlabeled samples via memory-efficient offline distillation, pushing the retention to 95. 16\% within only 30 minutes on a single GPU. Code is available at \url{https: //github. com/chenxinrui-tsinghua/LinearPatch}.

PDF Details

EAAI Journal 2025 Journal Article

Deep learning methods for protein representation and function prediction: A comprehensive overview

Mingqing Wang
Zhiwei Nie
Yonghong He
Athanasios V. Vasilakos
Qiang (Shawn) Cheng
Zhixiang Ren

Deep learning has revolutionized protein function prediction by capturing intricate protein relationships, yet a comprehensive survey of its methodologies remains elusive. In this review, we systematically dissect recent advances by addressing three pivotal questions: (a) which modalities are most critical for various function prediction tasks, (b) which deep learning strategies optimally model these modalities, (c) what common and task-specific challenges persist. We categorize protein data into eight distinct types – from fundamental representations to specialized expert knowledge – and provide an exhaustive analysis of state-of-the-art deep learning models alongside emerging self-supervised learning strategies. Moreover, we compare the evolution of architectures across different modeling paradigms, highlighting their respective strengths and limitations. Our investigation spans over fifteen downstream tasks across five key research areas, including protein function annotation, protein–protein interactions, protein–ligand interactions, mutation effect prediction, and remote homology detection. Finally, we discuss current challenges and propose potential solutions, offering strategic guidance for data selection, methodological innovation, and future research directions in the application of deep learning to protein function prediction.

Details DOI

JBHI Journal 2025 Journal Article

Hierarchically Optimized Multiple Instance Learning With Multi-Magnification Pathological Images for Cerebral Tumor Diagnosis

Lianghui Zhu
Renao Yan
Tian Guan
Fenfen Zhang
Linlang Guo
Qiming He
Shanshan Shi
Huijuan Shi

Accurate diagnosis of cerebral tumors is crucial for effective clinical therapeutics and prognosis. However, limitations in brain biopsy tissues and the scarcity of pathologists specializing in cerebral tumors hinder comprehensive clinical tests for precise diagnosis. To address these challenges, we first established a brain tumor dataset of 3, 520 cases collected from multiple centers. We then proposed a novel Hierarchically Optimized Multiple Instance Learning (HOMIL) method for classifying six common brain tumor types, glioma grading, and predicting the origin of brain metastatic cancers. The feature encoder and aggregator in HOMIL were trained alternately based on specific datasets and tasks. Compared to other multiple instance learning (MIL) methods, HOMIL achieved state-of-the-art performance with impressive accuracies: 93. 29% / 85. 60% for brain tumor classification, 91. 21% / 96. 93% for glioma grading, and 86. 36% / 79. 28% for origin determination on internal/external datasets. Additionally, HOMIL effectively located multi-scale regions of interest, enabling an in-depth analysis through features and heatmaps. Extensive visualization demonstrated HOMIL's ability to cluster features within the same type while establishing distinct boundaries between tumor types. It also identified critical areas on pathological slides, regardless of tumor size.

Details DOI

NeurIPS Conference 2023 Conference Paper

TexQ: Zero-shot Network Quantization with Texture Feature Distribution Calibration

Xinrui Chen
Yizhi Wang
Renao Yan
Yiqing Liu
Tian Guan
Yonghong He

Quantization is an effective way to compress neural networks. By reducing the bit width of the parameters, the processing efficiency of neural network models at edge devices can be notably improved. Most conventional quantization methods utilize real datasets to optimize quantization parameters and fine-tune. Due to the inevitable privacy and security issues of real samples, the existing real-data-driven methods are no longer applicable. Thus, a natural method is to introduce synthetic samples for zero-shot quantization (ZSQ). However, the conventional synthetic samples fail to retain the detailed texture feature distributions, which severely limits the knowledge transfer and performance of the quantized model. In this paper, a novel ZSQ method, TexQ is proposed to address this issue. We first synthesize a calibration image and extract its calibration center for each class with a texture feature energy distribution calibration method. Then, the calibration centers are used to guide the generator to synthesize samples. Finally, we introduce the mixup knowledge distillation module to diversify synthetic samples for fine-tuning. Extensive experiments on CIFAR10/100 and ImageNet show that TexQ is observed to perform state-of-the-art in ultra-low bit width quantization. For example, when ResNet-18 is quantized to 3-bit, TexQ achieves a 12. 18% top-1 accuracy increase on ImageNet compared to state-of-the-art methods. Code at https: //github. com/dangsingrue/TexQ.

PDF Details