290 citations · 1.3k across the 57 of their papers we have counts for
34 papers · 1 filter
Scaling Diffusion Transformers Efficiently via P
Chenyu Zheng, Xinyu Zhang, Rongzhen Wang +5
Diffusion Transformers have emerged as the foundation for vision generative models, but their scalability is limited by the high cost of hyperparameter (HP) tuning at large scales.…
On Mesa-Optimization in Autoregressively Trained Transformers: Emergence and Capability
Chenyu Zheng, Wei Huang, Rongzhen Wang +3
Autoregressively trained transformers have brought a profound revolution to the world, especially with their in-context learning (ICL) ability to address downstream tasks. Recently…
MissDiff: Training Diffusion Models on Tabular Data with Missing Values
Yidong Ouyang, Liyan Xie, Chongxuan Li +1
The diffusion model has shown remarkable performance in modeling data distributions and synthesizing data. However, the vanilla diffusion model requires complete or fully observed…
Towards Understanding Generalization of Macro-AUC in Multi-label Learning
Guoqiang Wu, Chongxuan Li, Yilong Yin
Macro-AUC is the arithmetic mean of the class-wise AUCs in multi-label learning and is commonly used in practice. However, its theoretical understanding is far lacking. Toward solv…
Contrastive Energy Prediction for Exact Energy-Guided Diffusion Sampling in Offline Reinforcement Learning
Cheng Lu, Huayu Chen, Jianfei Chen +3
Guided sampling is a vital approach for applying diffusion models in real-world tasks that embeds human-defined guidance during the sampling procedure. This paper considers a gener…
Toward Understanding Generative Data Augmentation
Chenyu Zheng, Guoqiang Wu, Chongxuan Li
Generative data augmentation, which scales datasets by obtaining fake labeled examples from a trained conditional generative model, boosts classification performance in various lea…
ProlificDreamer: High-Fidelity and Diverse Text-to-3D Generation with Variational Score Distillation
Zhengyi Wang, Cheng Lu, Yikai Wang +4
Score distillation sampling (SDS) has shown great promise in text-to-3D generation by distilling pretrained large-scale text-to-image diffusion models, but suffers from over-satura…
One Transformer Fits All Distributions in Multi-Modal Diffusion at Scale
Fan Bao, Shen Nie, Kaiwen Xue +7
This paper proposes a unified diffusion framework (dubbed UniDiffuser) to fit all distributions relevant to a set of multi-modal data in one model. Our key insight is -- learning d…
Revisiting Discriminative vs. Generative Classifiers: Theory and Implications
Chenyu Zheng, Guoqiang Wu, Fan Bao +3
A large-scale deep model pre-trained on massive labeled or unlabeled data transfers well to downstream tasks. Linear evaluation freezes parameters in the pre-trained model and trai…
Why Are Conditional Generative Models Better Than Unconditional Ones?
Fan Bao, Chongxuan Li, Jiacheng Sun +1
Extensive empirical evidence demonstrates that conditional generative models are easier to train and perform better than unconditional ones by exploiting the labels of data. So do…
DPM-Solver++: Fast Solver for Guided Sampling of Diffusion Probabilistic Models
Cheng Lu, Yuhao Zhou, Fan Bao +3
Diffusion probabilistic models (DPMs) have achieved impressive success in high-resolution image synthesis, especially in recent large-scale text-to-image generation applications. A…
Deep Generative Modeling on Limited Data with Regularization by Nontransferable Pre-trained Models
Yong Zhong, Hongtao Liu, Xiaodong Liu +3
Deep generative models (DGMs) are data-eager because learning a complex model on limited data suffers from a large variance and easily overfits. Inspired by the classical perspecti…
Fast Lossless Neural Compression with Integer-Only Discrete Flows
Siyu Wang, Jianfei Chen, Chongxuan Li +2
By applying entropy codecs with learned data distributions, neural compressors have significantly outperformed traditional codecs in terms of compression ratio. However, the high i…
Estimating the Optimal Covariance with Imperfect Mean in Diffusion Probabilistic Models
Fan Bao, Chongxuan Li, Jiacheng Sun +2
Diffusion probabilistic models (DPMs) are a class of powerful deep generative models (DGMs). Despite their success, the iterative generation process over the full timesteps is much…
DPM-Solver: A Fast ODE Solver for Diffusion Probabilistic Model Sampling in Around 10 Steps
Cheng Lu, Yuhao Zhou, Fan Bao +3
Diffusion probabilistic models (DPMs) are emerging powerful generative models. Despite their high-quality generation performance, DPMs still suffer from their slow sampling as they…
Memory Replay with Data Compression for Continual Learning
Liyuan Wang, Xingxing Zhang, Kuo Yang +7
Continual learning needs to overcome catastrophic forgetting of the past. Memory replay of representative old training samples has been shown as an effective solution, and achieves…
Analytic-DPM: an Analytic Estimate of the Optimal Reverse Variance in Diffusion Probabilistic Models
Fan Bao, Chongxuan Li, Jun Zhu +1
Diffusion probabilistic models (DPMs) represent a class of powerful generative models. Despite their success, the inference of DPMs is expensive since it generally needs to iterate…
Stability and Generalization of Bilevel Programming in Hyperparameter Optimization
Fan Bao, Guoqiang Wu, Chongxuan Li +2
The (gradient-based) bilevel programming framework is widely used in hyperparameter optimization and has achieved excellent performance empirically. Previous theoretical work mainl…
Rethinking and Reweighting the Univariate Losses for Multi-Label Ranking: Consistency and Generalization
Guoqiang Wu, Chongxuan Li, Kun Xu +1
(Partial) ranking loss is a commonly used evaluation measure for multi-label classification, which is usually optimized with convex surrogates for computational efficiency. Prior t…
MiCE: Mixture of Contrastive Experts for Unsupervised Image Clustering
Tsung Wei Tsai, Chongxuan Li, Jun Zhu
We present Mixture of Contrastive Experts (MiCE), a unified probabilistic clustering framework that simultaneously exploits the discriminative representations learned by contrastiv…
ORDisCo: Effective and Efficient Usage of Incremental Unlabeled Data for Semi-supervised Continual Learning
Liyuan Wang, Kuo Yang, Chongxuan Li +3
Continual learning usually assumes the incoming data are fully labeled, which might not be applicable in real applications. In this work, we consider semi-supervised continual lear…
Bi-level Score Matching for Learning Energy-based Latent Variable Models
Fan Bao, Chongxuan Li, Kun Xu +3
Score matching (SM) provides a compelling approach to learn energy-based models (EBMs) by avoiding the calculation of partition function. However, it remains largely open to learn…
Variational (Gradient) Estimate of the Score Function in Energy-based Latent Variable Models
Fan Bao, Kun Xu, Chongxuan Li +3
The learning and evaluation of energy-based latent variable models (EBLVMs) without any structural assumptions are highly challenging, because the true posteriors and the partition…
Efficient Learning of Generative Models via Finite-Difference Score Matching
Tianyu Pang, Kun Xu, Chongxuan Li +3
Several machine learning applications involve the optimization of higher-order derivatives (e.g., gradients of gradients) during training, which can be expensive in respect to memo…