Hello, I'm Anh Tien Nguyen

I am a 2nd-year CS PhD student at Stony Brook University. My works focus on mechanistic interpretability, sparse autoencoder, and computer vision. Previously, I have been working on computational pathology. Email: atinguyen (at) cs (dot) stonybrook (dot) edu

CV Scholar Github Twitter

News

Publications

Separating interpretability and reconstruction objectives in Sparse Autoencoders

Separating interpretability and reconstruction objectives in Sparse Autoencoders

Anh Tien Nguyen, Dimitris Samaras, Mahdi S. Hosseini
Under review

A method to separate interpretability and reconstruction objectives in sparse autoencoders

MonoLoss: A Training Objective for Interpretable Monosemantic Representations

MonoLoss: A Training Objective for Interpretable Monosemantic Representations

Ali Nasiri-Sarvi*, Anh Tien Nguyen*, Hassan Rivaz, Dimitris Samaras, Mahdi S. Hosseini
WACV 2027

A training objective to directly optimize monosemanticity

PDF Code
Normal and Abnormal Pathology Knowledge-Augmented Vision-Language Model for Anomaly Detection in Pathology Images

Normal and Abnormal Pathology Knowledge-Augmented Vision-Language Model for Anomaly Detection in Pathology Images

Jinsol Song, Jiamu Wang, Anh Tien Nguyen, Keunho Byeon, Sangjeong Ahn, SungHak Lee, Jin Tae Kwak
ICCV 2025

A model that detects anomalous pathology by utilizing expert knowledge of normal and abnormal histopathology as well as knowledge from a pre-trained VLM

PDF Code
Pathology-Informed Latent Diffusion Model for Anomaly Detection in Lymph Node Metastasis

Pathology-Informed Latent Diffusion Model for Anomaly Detection in Lymph Node Metastasis

Jiamu Wang, Keunho Byeon, Jinsol Song, Anh Tien Nguyen, Sangjeong Ahn, SungHak Lee, Jin Tae Kwak
MICCAI 2025

A pathology-informed LDM for unsupervised anomaly detection in lymph nodes

PDF Code
VLEER: Vision and Language Embeddings for Explainable Whole Slide Image Representation

VLEER: Vision and Language Embeddings for Explainable Whole Slide Image Representation

Anh Tien Nguyen, Keunho Byeon, Kyungeun Kim, Jin Tae Kwak
MICCAI 2025

A novel method designed to leverage VLMs for WSI representation with explainability

PDF
2DMamba: Efficient State Space Model for Image Representation with Applications on Giga-Pixel Whole Slide Image Classification

2DMamba: Efficient State Space Model for Image Representation with Applications on Giga-Pixel Whole Slide Image Classification

Jingwei Zhang*, Anh Tien Nguyen*, Xi Han*, Vincent Quoc-Huy Trinh, Hong Qin, Dimitris Samaras, Mahdi S. Hosseini
CVPR 2025

A novel 2D selective SSM framework that incorporates the 2D spatial structure of images into Mamba, with a highly optimized hardware-aware operator.

PDF Code
Towards a text-based quantitative and explainable histopathology image analysis

Towards a text-based quantitative and explainable histopathology image analysis

Anh Tien Nguyen, Trinh Thi Le Vuong, Jin Tae Kwak
MICCAI 2024 (Early accept, top 11%)

An explainable method to utilize texts for quantifying pathology images

PDF Code
CAMP: Continuous and Adaptive Learning Model in Pathology

CAMP: Continuous and Adaptive Learning Model in Pathology

Anh Tien Nguyen, Keunho Byeon, Kyungeun Kim, Boram Song, Seoung Wan, Jin Tae Kwak
npj Artificial Intelligence

A framework for continuous learning that is applicable for any vision model.

PDF Code
GPC: Generative and General Pathology Image Classifier

GPC: Generative and General Pathology Image Classifier

Anh Tien Nguyen, Jin Tae Kwak
MICCAI 2023 (Best Paper Honorable Mention)

A task-agnostic generative and general pathology image classifier

PDF Code