Unsolicited Opinions on AGI, RSI and Financial Markets (2026)
Resources
Open-source code, datasets, and tools for advancing machine learning and computer vision research.
Explore our curated collection of research artifacts, including reproducible code, pre-trained models, and benchmark datasets—freely available to support collaboration and innovation in the broader research community.
Selected Tutorials
Learn More
Where Does Learning Live in the Spectrum? (2026)
From Learning to Optimize to Learning Not to Optimize (2026)
What Holds Up a Graph Foundation Model? (2026)
Low Rank Strikes Back In Modern Foundation Models (2025)
Sparse Neural Networks: From Practice to Theory (2023)
Open Calls for Papers/Participation
NeurIPS Workshop on AI-Native Academia: Authorship, Peer Review, and Conference Governance under AI, Dec 2026
VITA in the Press (selected)
AI Street: "Why AI Agents Age", Jun 2026
Nature: "Too Much Social Media Gives AI Chatbots Brain Rot", Oct 2025
WIRED: "AI Models Get Brain Rot, Too", Oct 2025
Forbes: "Junky Online Content Gives AI Models Brain Rot Too", Oct 2025
IEEE Intelligent Systems: "AI’s 10 to Watch, 2022", Apr 2023
MIT News: "Learning to grow machine-learning models", Mar 2023
IEEE Computer Magazine: "Virtual Roundtable Interview: Efficient Computer Vision for Embedded Systems", Apr 2022
NSF News: "Accelerating Neural Architecture Search with Theory-Grounded, Training-Free Metrics", Mar 2022
Quanta Magazine: "Will Transformers Take Over Artificial Intelligence?", Mar 2022
MIT News: "Shrinking massive neural networks used to model language", Dec 2020
US Army scientists pick the top 10 coolest advances of 2020
Datasets
MIA'24, Long-tailed, multi-label, and zero-shot classification on chest X-rays (CXR-LT) dataset
IEEE TIP'23, Multi-Purpose Real Haze Benchmark (A2I2-Haze) dataset
IEEE TPAMI'20 + ECCV'18, Privacy attributes in HMDB51 (PA-HMDB51) dataset
IJCV'20 + CVPR'19, Multi-Purpose Image Deraining (MPID) dataset
IEEE TIP'19, REalistic Single Image DEhazing (RESIDE) dataset
ACM MM'15, Adobe Visual Font Recognition (VFR) dataset
Courses
At UT Austin
Spring 2024
Fall 2023
Spring 2023
Fall 2022
Spring 2022
Fall 2021
Spring 2021
At TAMU
Spring 2020
Fall 2019
Summer 2019
Advanced Summer School of Learning from Data
(Invited short course at University of Sao Paulo, Brazil)
Spring 2019
Fall 2018
Summer 2018
Recent Advances of Deep Learning in Computer Vision
(Invited short course at University of Sao Paulo, Brazil)
Spring 2018
Fall 2017
Codes
Other Softwares
-
APOLLO (MLSys'25) has been integrated into HuggingFace, LLaMA-Factory, Axolotl, and FluxML
-
Q-GaLore (CPAL'25) has been integrated into ModelScope SWIFT
-
GaLore (ICML'24) has been integrated into HuggingFace, PyTorch, LLaMA-Factory, FedML, Axolotl, and bitsandbytes
-
H2O (NeurIPS'23) has been integrated into DeepSpeed by Microsoft, and Llama-Recipes by Meta
-
Our three popular image enhancement algorithms: AOD-Net, EnlightenGAN, and DeblurGAN-V2, are included into the open-source GNU Image Manipulation Program toolbox (GIMP-ML), as deep-dehazing, enlighten, and deblur plugins
-
DeblurGAN-V2 is also implemented by Intel OpenVINO toolkit
-
DeepFont, an Adobe Photoshop built-in feature, as one of its leading contributors. [Official Photoshop Manual]
-
Distributed Machine Learning (DML) ToolBox, as part of Microsoft Prajna
-
Dehazing algorithms integrated in the Computer Vision Emergency Response (CVER) toolkit [by Pritish Uplavikar and Ryan Wells] [Webpage]