News
If you are here to seek "TL;DR"...
- Question: What's a quick read to get a snapshot of VITA's current research focus?
- Short Answer: We pick five papers that represent our recent flavors as "chef's choice". The list will change over time: [Large Spatial Model, NeurIPS 2024], [GaLore, ICML 2024], [Flextron, ICML 2024], [InstantSplat, arXiv 2024], [Heavy-Hitter Oracle, NeurIPS 2023]
- Long Answer: Please refer to our (still brief) Research Agenda. Please also consider
- Since May 2024, Dr. Wang has been on leave from UT Austin to serve as the full-time Research Director for XTX Markets. You can read more [here]
[Oct. 2024]
[Sep. 2024]
- 8 NeurIPS'24 (LightGaussian + expressive gaussian avatar + Read-ME + Found in the Middle + Large Spatial Model + transformer training dynamics + Diffusion4D + AlphaPruning) accepted
- 1 NeurIPS Datasets & Benchmark Track'24 (Model-GLUE) accepted
- 1 IEEE Trans. PAMI (symbolic visual RL) accepted
- Our group co-organized the ECCV 2024 "Sometimes Less is More: the 1st Dataset Distillation Challenge"
[Aug. 2024]
- We are grateful to receive the Best Paper Finalist Award from VLDB 2024 [Paper]
- Dr. Wang is grateful to receive the AI 100: The Top AI Thought Leaders award, presented by H2O.ai
- 1 JMLR (pruning provably improves generalization) accepted
- 1 JMLR (tighter theoretical analysis of sparse activation) accepted
- VITA welcomes three new Ph.D. students: Junbo Li, Yuehao Wang, and Kevin Wang
[Jul. 2024]
- 3 ECCV'24 (Few-shot 3DGS + DreamScene360 + VersatileGaussian) accepted
- 1 npj Digital Medicine (longitudinal medical imaging) accepted
[Jun. 2024]
[May. 2024]
- VITA's Summer 2024 squad of Ph.D. students graduated:
- Drs. Yifan Jiang and Scott Hoang both joined Apple AIML as Senior Research Scientists (both were super excited about their packages!)
- Dr. Yuning You will join CS@CUHK Shenzhen as an Assistant Professor (starting from Fall 2025), after another year of postdoc at Caltech
Congratulations to ALL!
- VITA Postdoc Junyuan Hong is selected to receive the ML and Systems Rising Star Award 2024
- 10 ICML'24 (GaLore + Flextron + long convolution LLM + LLM junk DNA + OWL pruning + LLaGA + decoding compressed trust + sparse cocktail + LESS + zeroth-order training) accepted
- 1 ACL'24 (LLM model soup) accepted
- 1 Medical Image Analysis (multi-label long-tail chest X-ray) accepted
- Our group co-organized the ICLR 2024 Workshop on Data-centric Machine Learning Research (DMLR): Harnessing Momentum for Science
[Mar. 2024]
- 1 IEEE Trans. PAMI (zero-shot NAS) accepted
- 1 WWW'24 (dynamic graph training) accepted
[Feb. 2024]
- 7 CVPR'24 (mode collapse in text-to-3D + Lift3D + Feature 3DGS + PAIR Diffusion + Zero-Painter + Prompt-Free Diffusion + OpenBias) accepted
- 2 MLSys'24 (Q-Hitter + formal verification LLM tuning) accepted
- 1 JAMA Cardiology (multimodal AS biomarker) accepted
[Jan. 2024]
- Our group co-organized the inaugural Conference on Parsimony and Learning (CPAL), Hong Kong
- 10 ICLR'24 (DP-OPT + LLM-KICK benchmark + Efficient-3DiM + hyperparameter scaling + DeepSet dimension + "vodka" distillation + graph diffusion + social reward + efficient watermarking + reweighted robust training) accepted
- 1 JAMIA (data-efficient electrocardiographic learning) accepted
[Dec. 2023]
- 1 SatML'24 (generative model privacy) accepted
- 1 TMLR (MoE in contrastive learning) accepted
[Nov. 2023]
- 1 IEEE Trans. CAS-II (neuro-symbolic HW-SW co-design) accepted
- 4 CPAL'24 (block sparse training + sparsity in material science + GAN micromotion + WS-iFSD) accepted
[Oct. 2023]
[Sep. 2023]
- 7 NeurIPS'23 (essential sparsity + LLM heavy-hitter oracle + pruning topology + graph MoE + in-context diffusion + patch diffusion + channel-level DST) accepted
- VITA Postdoc Shiwei Liu is selected to receive the prestigious Newton International Fellowship from the British Academy and the Royal Society
[Aug. 2023]
- VITA welcomes three new Ph.D. students: Wes Robbins, Seo Young Lee, and Runjin Chen
- 1 Nature Communications (medical imaging fairness) accepted
- 1 IEEE Trans. Image Processing (adaptive deblurring) accepted
[Jul. 2023]
- 8 ICCV'23 (GNT-MOVE + turbulence diffusion + vision HGNN + AdaMV-MoE + StegaNeRF + versatile diffusion + robust MoE + Text2Video-Zero) accepted
- 1 QCE'23 (quantum sparse training) accepted
- 2 ICCAD'23 (Edge-MoE + INR dataflow) accepted
- 1 invited paper at IEEE Data Engineering Bulletin (graph contrastive learning)
- VITA Ph.D. students Yifan Jiang and Greg Holste are selected to receive two UT Graduate Dean’s Prestigious Fellowships
[Jun. 2023]
[May. 2023]
- VITA's Summer 2023 (big) squad of Ph.D. students graduated:
- Dr. Tianlong Chen will join CS@UNC Chapel Hill as an Assistant Professor (starting from Fall 2024), after another year of postdoc at MIT/Harvard
- Dr. Wuyang Chen will join CS@Simon Fraser as an Assistant Professor (starting from Fall 2024), after another year of postdoc at UC Berkeley
- Dr. Junru Wu joins Google Deepmind, New York as a Research Engineer
- Dr. Haotao Wang joins Qualcomm AI Research, San Diego as a Research Scientist
- Dr. Ziyu Jiang joins NEC Labs America, CA as a Research Scientist
- Dr. Jiayi Shen joins Amazon, CA as an Applied Scientist
- Dr. Yan Han joins LinkedIn, CA as a Research Scientist
Congratulations to ALL!
- Announce the Conference on Parsimony and Learning (CPAL), a brand-new conference for the sparse ML research community, that Dr. Wang co-founds and will serve as its inaugural Program Chair
- 1 ACL'23 (sparse LLM tuning) accepted
- 1 MICCAI'23 (multi-label long-tail chest X-ray) accepted
- 1 AutoML-Conf'23 (neural architecture "no free lunch") accepted
- 1 TMLR (partial graph transfer learning) accepted
- Our group co-organized the ICLR 2023 Workshop on Sparsity in Neural Networks: On Practical Limitations and Tradeoffs between Sustainability and Efficiency
[Apr. 2023]
- 9 ICML'23 (Instant soup + graph ladling + robust weight signature + neural scaling law + code generation + L2O game + large kernel teacher + pre-training tax + L2O math structure) accepted
- Dr. Wang is grateful to receive the 2023 Google Research Scholar Award
- Our group co-organized the NSF IFML workshop on Models v.s. Data in ML and AI
[Mar. 2023]
[Feb. 2023]
- 4 CVPR'23 (NeuralLift-360 + AligNeRF + MMG-Ego4D + specialist diffusion) accepted
- 1 TMLR (robust fairness) accepted
- Dr. Wang is grateful to receive tenure promotion (effective Sep 2023)
[Jan. 2023]
- 14 ICLR'23 (Ramanujan graph pruning + SMC benchmark + MoE as dropout + learning to grow + federated OoD + ConvNet 51x51 + GNT + NeRF-SOS + layered pre-training + HotProtein + hypergraph diffusion + graph domain adaptation + L2O adaptation + protein convolution) accepted
- 2 AISTATS'23 (sparse network NTK + L2O generalization) accepted
- 1 IEEE Trans. Image Processing (A2I2 Real Haze benchmark) accepted
- We thank MattVidPro AI for the very cool video [Youtube] highlighting our latest work, Versatile Diffusion (ICCV'23) [Paper] [Code] [Hugging Face]
- We thank IEEE Signal Processing Society [Link] for highlighting our IEEE TIP'21 work, EnlightenGAN [Paper] [Code], as one of SPS's top-25 most downloaded articles on IEEE Xplore®, 2021-2022
- Our group co-organized the 3rd Workshop on Seeking Low-Dimensionality in Deep Neural Networks (SLowDNN)
[Dec. 2022]
- We are grateful to receive the Best Paper Award from LoG 2022 [Paper]
- Dr. Wang is grateful to receive the IEEE biennial "AI's 10 To Watch" award in 2022
- VITA Ph.D. student Yifan Jiang is selected to receive a 2022 academic year Apple Scholars in AI/ML PhD fellowship
[Nov. 2022]
- VITA Ph.D. student Dejia Xu is selected to receive a 2022 academic year Snap Research Fellowship
- Our group is selected to be supported by the Meta Reality Labs Research Gift Award
- Our group is selected to be supported by the Google TensorFlow Model Garden Award (again)
- 3 AAAI'23 (safeguarded L2O + federated robustness + efficient ViT) accepted
- 1 LoG'22 (untrained graph LTH) accepted
[Oct. 2022]
- 1 WSDM'23 (search behavior prediction) accepted
- 1 WACV'23 (imbalance in medical image localization) accepted
[Sep. 2022]
- 11 NeurIPS'22 (DAG convergence + INR DSP theory + M3ViT + back razor + randomized channel shuffling + symbolic TCC + GCN gradient + trapping backdoor + multimodal training + sparse few-shot learning + hypergraph contrastive learning) accepted
- 1 NeurIPS Datasets & Benchmark Track'22 (large-scale graph training) accepted
[Aug. 2022]
- VITA Ph.D. student Zhiwen Fan is selected to receive a 2022 academic year Qualcomm Innovation Fellowship (QIF)
- VITA welcomes four new Ph.D. students: Zhenyu Zhang, Tong Wang, Ruisi Cai, and Wenyan Cong
- 1 IEEE Trans. Medical Imaging (contrastive learning on X-rays) accepted
- 1 TMLR (all-pass lottery) accepted
- 1 ICDM'22 (stochastic KD for medical imaging) accepted
- 1 CIKM'22 (multi-modal recommendation) accepted
- Our group co-organized MLSys 2022 Tutorial: Training-Free Approaches for Edge AI: Challenges, Opportunities and Progress
[Jul. 2022]
- 8 ECCV'22 (SinNeRF + INR stylization + scalable L2O + point cloud MAE + few-shot align + malleable convolution + universal ViT + turbulence) accepted
- 1 TMLR (adversarial feature augmentation) accepted
- VITA Ph.D. student Xiaohan Chen graduated and joined Alibaba Damo Academy (Decision Intelligence Lab), Seattle, as a full-time senior research scientist ("Ali Star" hire)
- Our group co-organized the 2nd workshop on Sparsity in Neural Networks: Advancing Understanding and Practice (SNN)
[Jun. 2022]
- Dr. Wang is grateful to receive the NSF CAREER Award
- Dr. Wang is grateful to receive the Aharon Katzir Young Investigator Award of International Neural Network Society (INNS)
- 1 ACM Computing Surveys (ML safety) accepted
- 1 TMLR (robust lifelong learning) accepted
- 3 ACM MM'22 (controllable light enhancement + video action detection + Cloud2Sketch) accepted
- 1 ACM BCB'22 (pretraining for COVID-19 prediction) accepted
- 1 MLHC'22 (personalized imbalanced training) accepted
- Our group co-organized the CVPR 2022 Workshop and Challenge on Bridging Computational Photography and Visual Recognition (UG2+)
- Our group co-organized the AICAS 2022 tutorial: IEEE Low-Power Computer Vision Challenge. A similar tutorial would be offered in DAC 2022
[May. 2022]
- 9 ICML'22 (long-tail OoD detection + BN-free robust training + neural implicit dictionary + improved sparse training + linearity grafting + structured lottery ticket + double-win lottery ticket + renormalization group theory + NN architecture growing) accepted
- 1 AutoML-Conf'22 (GNN NAS) accepted
[Apr. 2022]
- 1 JMLR (learning to optimize) accepted
- 1 IEEE Trans. PAMI (deep GCN training benchmark) accepted
- 1 ACM FAccT'22 (data efficiency under differential privacy) accepted
- We thank IEEE Computer Magazine [article] for covering our recent success in building energy-efficient computer vision systems
[Mar. 2022]
- We thank Quanta Magazine [article] for covering our NeurIPS'21 work, TransGAN
- We thank National Science Foundation (NSF) news [article] for covering our training-free NAS works, in ICLR'21 (TE-NAS) and ICLR'22 (As-ViT)
[Feb. 2022]
- 7 CVPR'22 (ViT training + Aug-NeRF + symbolic spotting + sparsity for Trojan + sparse multi-tasking + video SR + fashion CLIP) accepted
- 1 IEEE Trans. Image Processing (light enhancement with noise) accepted
[Jan. 2022]
- 14 ICLR'22 (symbolic L2O + disguised subnetwork + optimizer amalgamation + robust sparsity + Fourier ViT + auto-scaling ViT + ViT compression + Frank-Wolfe pruning + cold brew + audio lottery + split-max + sparse ensemble + random sparse training + Bayesian L2O) accepted
- 1 AISTATS'22 (variational feature selection) accepted
- 2 ICASSP'22 (lifelong speech synthesis + sensor data imputation) accepted
- 1 ACM Trans. DAES (efficient segmentation) accepted
- VITA Ph.D. student Tianlong Chen is selected to receive a 2022 academic year Adobe PhD Research Fellowship - that is after him being awarded IBM PhD Fellowship 2021, UT Graduate Dean’s Prestigious Fellowship 2021, and Baidu Scholarship finalist
- VITA welcomes four new Ph.D. students: Xuxi Chen, Dejia Xu, Hongru Yang and Zhangheng Li
[Dec. 2021]
- People in this group did nothing this month but enjoyed some well-deserved vacation time, with their families and loved ones
[Nov. 2021]
[Oct. 2021]
- 1 WSDM'22 (graph contrastive learning) accepted
- 1 IEEE Trans. SPIN (vision-based drone swarm control) accepted
- 3 WACV'22 (video NAS + sandwich BatchNorm + chest X-rays) accepted
- Our group is selected to be supported by the Google TensorFlow Model Garden Award
- Our group co-organized the ICCV 2021 Workshop on Real-world Recognition from Low-quality Inputs (RLQ)
- VITA Postdoc Guoliang Kang joins the CS department of University of Technology Sydney, Australia, as Lecturer (Assistant Professor) and endowed with the prestigious DECRA Fellow
- We thank Henry for the very cool video [Youtube] highlighting our latest work, AugMax (NeurIPS'21) [Paper] [Code]
[Sep. 2021]
- 13 NeurIPS'21 (TransGAN + AugMax + data-efficient GAN + elastic lottery ticket + SViTE + DePT + stealing lottery + imbalanced contrastive learning + HyperLISTA + WeakNAS + IA-RED2 + neuroregeneration + lottery ticket benchmark) accepted
- Our group won the 1st prize of IEEE 2021 Low-Power Computer Vision Challenge (video track) [Solution] [Tech Report]
- 1 ICCV'21 workshop (graph CNN for motion prediction) accepted
[Aug. 2021]
- VITA welcomes five new Ph.D. students: Zhiwen Fan, Scott Hoang, Peihao Wang, Greg Holste, and Ajay Jaiswal
- 1 IEEE Trans. Image Processing (sketch-to-image synthesis) accepted
- 1 Springer Machine Learning (weakly-supervised segmentation troubleshooting) accepted
[Jul. 2021]
[Jun. 2021]
[May. 2021]
- VITA's Summer 2021 squad of students graduated: Dr. Zhenyu Wu joined Wormpex AI Research (Seattle) as Research Scientist, Scott continues as VITA Ph.D. student, and Rahul joined ByteDance AI Lab (Silicon Valley) as Machine Learning Engineer. Congratulations!
- 5 ICML'21 (graph contrastive learning + homotopy attack + imbalanced contrastive learning + graph lottery ticket + data-efficient lottery ticket) accepted
- 1 KDD'21 (federated learning debiasing) accepted
- 1 ACL'21 (EarlyBERT) accepted
- Our group is selected to be supported by the NVIDIA Applied Research Accelerator Program
[Apr. 2021]
- VITA Ph.D. student Tianlong Chen is selected to receive a 2021 academic year IBM PhD Fellowship -- that is after Tianlong being selected as a Baidu Scholarship 2021 finalist
- 2 CVPR'21 workshop (CNN high-frequency bias + BN-free training of binary networks) accepted
[Mar. 2021]
- 1 ICME'21 (arbitrary style transfer) accepted
- We thank UT Engineering News [article] for highlighting our group's research
[Feb. 2021]
- 3 CVPR'21 (CV lottery ticket + blind image IQA + assessing image enhancement) accepted
- 1 IJCV (open-world generalization of segmentation models) accepted
- 1 DAC'21 + 1 ICASSP'21 accepted (InstantNet + VGAI)
- We thank Yannic for the very cool video [Youtube] highlighting our latest work, TransGAN (NeurIPS'21) [Paper] [Code]
[Jan. 2021]
- 8 ICLR'21 (lifelong lottery ticket + robust overfitting + nasty knowledge distillation + theory-guided NAS + domain generalization + LISTA unrolling + learning to optimize for minimax + efficient recommendation system) accepted
- 1 IEEE Trans. PAMI (artistic text style transfer) accepted
- We thank IDG Connect [article] for covering our ICLR'20 work, Early Bird Lottery Ticket for Effcient Deep Learning
- Our group co-organized the IJCAI 2020 BOOM Workshop
- VITA welcomes new Ph.D. student Wenqing Zheng
[Dec. 2020]
[Nov. 2020]
[Oct. 2020]
- Our three popular image enhancement algorithms: AOD-Net (ICCV 2017), EnlightenGAN (IEEE TIP 2020), and DeblurGAN-V2 (ICCV 2019), are included into the open-source GNU Image Manipulation Program toolbox (GIMP-ML), as deep-dehazing, enlighten, and deblur plugins
[Sep. 2020]
- Dr. Wang is grateful to receive the 2020 Adobe Data Science Research Award
- 8 NeurIPS'20 (learning to optimize + once-for-all adversarial training + BERT lottery ticket + meta learning + robust contrastive learning + graph contrastive learning + ShiftAddNet + efficient quantized training) accepted
- 1 IEEE Trans. PAMI (privacy-preserving visual recognition) accepted
[Aug. 2020]
[Jul. 2020]
- 3 ECCV'20 (GAN compression + sketch-to-image synthesis + on-device learning-to-optimize) accepted
- 1 ACM Multimedia'20 (MMHand synthesizer) accepted
- 1 InterSpeech'20 (AutoSpeech) accepted
- We thank Tech Xplore [article] for covering our work, adversarial 3-D logos
[Jun. 2020]
- 6 ICML'20 (domain generalization + noisy label training + self-supervised GCN + DNN optimization + GAN compression + NAS for Bayesian models) accepted
- Our group co-organized the CVPR 2020 UG2+ Workshop and Prize Challenge
[May. 2020]
- 1 IEEE Trans. PAMI (image enhancement for visual understanding) accepted
- 1 IEEE Trans. Mobile Computing (adaptive model compression) accepted
[Apr. 2020]
- Dr. Wang is grateful to receive the 2020 ARO Young Investigator Award (YIP)
- Dr. Wang is grateful to receive the 2020 IBM Faculty Research Award
- Dr. Wang is grateful to receive the 2020 Amazon Research Award (AWS AI)
- 2 CVPR'20 workshop (efficient triplet loss + fine-grained classification) accepted
[Mar. 2020]
- 1 ISCA'20 (algorithm-hardware co-design to reduce data movement) accepted
[Feb. 2020]
- 3 CVPR'20 (self-supervised adversarial robustness + fast GCN training + indoor scene reasoning) accepted
- 1 IEEE Trans. Image Processing (visual understanding in poor-visibility environments) accepted
[Jan. 2020]
-
1 AISTATS'20 (CNN uncertainty quantification) accepted
- 1 IEEE Trans. CSVT (GAN data augmentation) accepted