论文检索

Deep Learning · Sequential Models, Time series

Shuai Zhang, Chuan Zhou, Yang Liu, PENG ZHANG, Xixun Lin, Zhiming Ma

We present a novel perspective on temporal point processes (TPPs) by reformulating their intensity processes as solutions to stochastic differential equations (SDEs). In particular, we first prove the equivalent SDE formulations of several classical TPPs, including Poisson processes, Hawkes processes, and self-correcting processes. Based on these proofs, we introduce a unified TPP framework called Neural Jump-Diffusion Temporal Point Process (NJDTPP), whose intensity process is governed by a neural jump-diffusion SDE (NJDSDE) where the drift, diffusion, and jump coefficient functions are parameterized by neural networks. Compared to previous works, NJDTPP exhibits model flexibility in capturing intensity dynamics without relying on any specific functional form, and provides theoretical guarantees regarding the existence and uniqueness of the solution to the proposed NJDSDE. Experiments on both synthetic and real-world datasets demonstrate that NJDTPP is capable of capturing the dynamics of intensity processes in different scenarios and significantly outperforms the state-of-the-art TPP models in prediction tasks.

Deep Learning · Generative Models and Autoencoders

Ziyi Zhang, Sen Zhang, Yibing Zhan, Yong Luo, Yonggang Wen, Dacheng Tao

Bridging the gap between diffusion models and human preferences is crucial for their integration into practical generative workflows. While optimizing downstream reward models has emerged as a promising alignment strategy, concerns arise regarding the risk of excessive optimization with learned reward models, which potentially compromises ground-truth performance. In this work, we confront the reward overoptimization problem in diffusion model alignment through the lenses of both inductive and primacy biases. We first identify a mismatch between current methods and the temporal inductive bias inherent in the multi-step denoising process of diffusion models, as a potential source of reward overoptimization. Then, we surprisingly discover that dormant neurons in our critic model act as a regularization against reward overoptimization while active neurons reflect primacy bias. Motivated by these observations, we propose Temporal Diffusion Policy Optimization with critic active neuron Reset (TDPO-R), a policy gradient algorithm that exploits the temporal inductive bias of diffusion models and mitigates the primacy bias stemming from active neurons. Empirical results demonstrate the superior efficacy of our methods in mitigating reward overoptimization. Code is avaliable at https://github.com/ZiyiZhang27/tdpo.

Weizhong Zhang, Zhiwei Zhang, Renjie Pi, Zhongming Jin, Yuan Gao, Jieping Ye, Kani Chen

Diffusion models have exhibited remarkable advancements in generating high-quality data. However, a critical drawback is their computationally intensive inference process, which requires a large number of timesteps to generate a single sample. Existing methods address this challenge by decoupling the forward and reverse processes, and they rely on handcrafted rules for sampling acceleration, leading to the risk of discarding important steps. In this paper, we propose an Efficient Denoising Diffusion method via Probabilistic Masking (EDDPM) that can identify and skip the redundant steps during training. To determine whether a timestep should be skipped or not, we employ probabilistic reparameterization to continualize the binary determination mask. The mask distribution parameters are learned jointly with model weights. By incorporating a real-time sparse constraint, our method can effectively identify and eliminate unnecessary steps during the training iterations, thereby improving inference efficiency. Notably, as the model becomes fully trained, the random masks converge to a sparse and deterministic one, retaining only a small number of essential steps. Empirical results demonstrate the superiority of our proposed EDDPM over the state-of-the-art sampling acceleration methods across various domains. EDDPM can generate high-quality samples with only 20% of the steps for time series imputation and achieve 4.89 FID with 5 steps for CIFAR-10. Moreover, when starting from a pretrained model, our method efficiently identifies the most informative timesteps within a single epoch, which demonstrates the potential of EDDPM to be a practical tool to explore large diffusion models with limited resources.

Theory · Learning Theory

Kaihong Zhang, Heqi Yin, Feng Liang, Jingbo Liu

We study the asymptotic error of score-based diffusion model sampling in large-sample scenarios from a non-parametric statistics perspective. We show that a kernel-based score estimator achieves an optimal mean square error of $\widetilde{O}\left(n^{-1} t^{-\frac{d+2}{2}}(t^{\frac{d}{2}} \vee 1)\right)$ for the score function of $p_0*\mathcal{N}(0,t\boldsymbol{I}_d)$, where $n$ and $d$ represent the sample size and the dimension, $t$ is bounded above and below by polynomials of $n$, and $p_0$ is an arbitrary sub-Gaussian distribution. As a consequence, this yields an $\widetilde{O}\left(n^{-1/2} t^{-\frac{d}{4}}\right)$ upper bound for the total variation error of the distribution of the sample generated by the diffusion model under a mere sub-Gaussian assumption. If in addition, $p_0$ belongs to the nonparametric family of the $\beta$-Sobolev space with $\beta\le 2$, by adopting an early stopping strategy, we obtain that the diffusion model is nearly (up to log factors) minimax optimal. This removes the crucial lower bound assumption on $p_0$ in previous proofs of the minimax optimality of the diffusion model for nonparametric families.

Deep Learning · Algorithms

Fangzhao Zhang, Mert Pilanci

Low-Rank Adaptation (LoRA) emerges as a popular parameter-efficient fine-tuning (PEFT) method, which proposes to freeze pretrained model weights and update an additive low-rank trainable matrix. In this work, we study the enhancement of LoRA training by introducing an $r\times r$ preconditioner in each gradient step where $r$ is the LoRA rank. We theoretically verify that the proposed preconditioner stabilizes feature learning with LoRA under infinite-width NN setting. Empirically, the implementation of this new preconditioner requires a small change to existing optimizer code and creates virtually minuscule storage and runtime overhead. Our experimental results with both large language models and text-to-image diffusion models show that with this new preconditioner, the convergence and reliability of SGD and AdamW can be significantly enhanced. Moreover, the training process becomes much more robust to hyperparameter choices such as learning rate. The new preconditioner can be derived from a novel Riemannian metric in low-rank matrix field.

Deep Learning · Generative Models and Autoencoders

Yahong Zhang, Chenchen Fan, Donghui Chen, Congrui Li, Wenli Ouyang, Mingda Zhu, Junchi Yan

Machine learning (ML) has been actively adopted in Linear Programming (LP) and Mixed-Integer Linear Programming (MILP), whose potential is hindered by instance scarcity. Current synthetic instance generation methods often fall short in closely mirroring the distribution of original datasets or ensuring the feasibility and boundedness of the generated data — a critical requirement for obtaining reliable supervised labels in model training. In this paper, we present a diffusion-based LP/MILP instance generative framework called MILP-FBGen. It strikes a balance between structural similarity and novelty while maintaining feasibility/boundedness via a meticulously designed structure-preserving generation module and a feasibility/boundedness-constrained sampling module. Our method shows superiority on two fronts: 1) preservation of key properties (hardness, feasibility, and boundedness) of LP/MILP instances, and 2) enhanced performance on downstream tasks. Extensive studies show two-fold superiority that our method ensures higher distributional similarity and 100% feasibility in both easy and hard datasets, surpassing current state-of-the-art techniques.

Deep Learning · Self-Supervised Learning

Zelin Zang, Hao Luo, Kai Wang, Panpan Zhang, Fan Wang, Stan Z Li, Yang You

Unsupervised Contrastive learning has gained prominence in fields such as vision, and biology, leveraging predefined positive/negative samples for representation learning. Data augmentation, categorized into hand-designed and model-based methods, has been identified as a crucial component for enhancing contrastive learning. However, hand-designed methods require human expertise in domain-specific data while sometimes distorting the meaning of the data. In contrast, generative model-based approaches usually require supervised or large-scale external data, which has become a bottleneck constraining model training in many domains. To address the problems presented above, this paper proposes DiffAug, a novel unsupervised contrastive learning technique with diffusion mode-based positive data generation. DiffAug consists of a semantic encoder and a conditional diffusion model; the conditional diffusion model generates new positive samples conditioned on the semantic encoding to serve the training of unsupervised contrast learning. With the help of iterative training of the semantic encoder and diffusion model, DiffAug improves the representation ability in an uninterrupted and unsupervised manner. Experimental evaluations show that DiffAug outperforms hand-designed and SOTA model-based augmentation methods on DNA sequence, visual, and bio-feature datasets. The code for review is released at [DiffAug CODE](https://github.com/zangzelin/code_diffaug).

Deep Learning · Generative Models and Autoencoders

Sunjae Yoon, Gwanhyeong Koo, Geonwoo Kim, Chang Yoo

In video editing, the hallmark of a quality edit lies in its consistent and unobtrusive adjustment. Modification, when integrated, must be smooth and subtle, preserving the natural flow and aligning seamlessly with the original vision. Therefore, our primary focus is on overcoming the current challenges in high quality edit to ensure that each edit enhances the final product without disrupting its intended essence. However, quality deterioration such as blurring and flickering is routinely observed in recent diffusion video editing systems. We confirm that this deterioration often stems from high-frequency leak: the diffusion model fails to accurately synthesize high-frequency components during denoising process. To this end, we devise Frequency Adapting Group (FRAG) which enhances the video quality in terms of consistency and fidelity by introducing a novel receptive field branch to preserve high-frequency components during the denoising process. FRAG is performed in a model-agnostic manner without additional training and validates the effectiveness on video editing benchmarks (i.e., TGVE, DAVIS).

Deep Learning · Generative Models and Autoencoders

Ling Yang, Zhaochen Yu, Chenlin Meng, Minkai Xu, Stefano Ermon, Bin Cui

Diffusion models have exhibit exceptional performance in text-to-image generation and editing. However, existing methods often face challenges when handling complex text prompts that involve multiple objects with multiple attributes and relationships. In this paper, we propose a brand new training-free text-to-image generation/editing framework, namely Recaption, Plan and Generate (RPG), harnessing the powerful chain-of-thought reasoning ability of multimodal LLMs to enhance the compositionality of text-to-image diffusion models. Our approach employs the MLLM as a global planner to decompose the process of generating complex images into multiple simpler generation tasks within subregions. We propose complementary regional diffusion to enable region-wise compositional generation. Furthermore, we integrate text-guided image generation and editing within the proposed RPG in a closed-loop fashion, thereby enhancing generalization ability. Extensive experiments demonstrate our RPG outperforms state-of-the-art text-to-image models, including DALL-E 3 and SDXL, particularly in multi-category object composition and text-image semantic alignment. Notably, our RPG framework exhibits wide compatibility with various MLLM architectures and diffusion backbones. Our code is available at https://github.com/YangLing0818/RPG-DiffusionMaster

Deep Learning · Large Language Models

Rui Yang, Xiaoman Pan, Feng Luo, Shuang Qiu, Han Zhong, Dong Yu, Jianshu Chen

We consider the problem of multi-objective alignment of foundation models with human preferences, which is a critical step towards helpful and harmless AI systems. However, it is generally costly and unstable to fine-tune large foundation models using reinforcement learning (RL), and the multi-dimensionality, heterogeneity, and conflicting nature of human preferences further complicate the alignment process. In this paper, we introduce Rewards-in-Context (RiC), which conditions the response of a foundation model on multiple rewards in its prompt context and applies supervised fine-tuning for alignment. The salient features of RiC are simplicity and adaptivity, as it only requires supervised fine-tuning of a single foundation model and supports dynamic adjustment for user preferences during inference time. Inspired by the analytical solution of an abstracted convex optimization problem, our dynamic inference-time adjustment method approaches the Pareto-optimal solution for multiple objectives. Empirical evidence demonstrates the efficacy of our method in aligning both Large Language Models (LLMs) and diffusion models to accommodate diverse rewards with only around 10% GPU hours compared with multi-objective RL baseline.

Deep Learning · Generative Models and Autoencoders

Lingxiao Yang, Shutong Ding, Yifan Cai, Jingyi Yu, Jingya Wang, Ye Shi

Recent advances in diffusion models attempt to handle conditional generative tasks by utilizing a differentiable loss function for guidance without the need for additional training. While these methods achieved certain success, they often compromise on sample quality and require small guidance step sizes, leading to longer sampling processes. This paper reveals that the fundamental issue lies in the manifold deviation during the sampling process when loss guidance is employed. We theoretically show the existence of manifold deviation by establishing a certain lower bound for the estimation error of the loss guidance. To mitigate this problem, we propose Diffusion with Spherical Gaussian constraint (DSG), drawing inspiration from the concentration phenomenon in high-dimensional Gaussian distributions. DSG effectively constrains the guidance step within the intermediate data manifold through optimization and enables the use of larger guidance steps. Furthermore, we present a closed-form solution for DSG denoising with the Spherical Gaussian constraint. Notably, DSG can seamlessly integrate as a plugin module within existing training-free conditional diffusion methods. Implementing DSG merely involves a few lines of additional code with almost no extra computational overhead, yet it leads to significant performance improvements. Comprehensive experimental results in various conditional generation tasks validate the superiority and adaptability of DSG in terms of both sample quality and time efficiency.

Applications · Time Series

Tijin Yan, Hengheng Gong, Yongping He, Yufeng Zhan, Yuanqing Xia

Probabilistic time series modeling based on generative models has attracted lots of attention because of its wide applications and excellent performance. However, existing state-of-the-art models, based on stochastic differential equation, not only struggle to determine the drift and diffusion coefficients during the design process but also have slow generation speed. To tackle this challenge, we firstly propose decomposable denoising diffusion model ($\text{D}^3\text{M}$) and prove it is a general framework unifying denoising diffusion models and continuous flow models. Based on the new framework, we propose some simple but efficient probability paths with high generation speed. Furthermore, we design a module that combines a special state space model with linear gated attention modules for sequence modeling. It preserves inductive bias and simultaneously models both local and global dependencies. Experimental results on 8 real-world datasets show that $\text{D}^3\text{M}$ reduces RMSE and CRPS by up to 4.6% and 4.3% compared with state-of-the-arts on imputation tasks, and achieves comparable results with state-of-the-arts on forecasting tasks with only 10 steps.

Applications · Language, Speech and Dialog

Manjie Xu, Chenxing Li, Duzhen Zhang, dan su, Wei Liang, Dong Yu

Audio editing involves the arbitrary manipulation of audio content through precise control. Although text-guided diffusion models have made significant advancements in text-to-audio generation, they still face challenges in finding a flexible and precise way to modify target events within an audio track. We present a novel approach, referred to as **PPAE**, which serves as a general module for diffusion models and enables precise audio editing. The editing is based on the input textual prompt only and is entirely training-free. We exploit the cross-attention maps of diffusion models to facilitate accurate local editing and employ a hierarchical local-global pipeline to ensure a smoother editing process. Experimental results highlight the effectiveness of our method in various editing tasks.

Deep Learning · Generative Models and Autoencoders

Jie Xiao, Kai Zhu, Han Zhang, Zhiheng Liu, Yujun Shen, Zhantao Yang, Ruili Feng, Yu Liu, Xueyang Fu, Zheng-Jun Zha

Consistency Models (CMs) have showed a promise in creating high-quality images with few steps. However, the way to add new conditional controls to the pre-trained CMs has not been explored. In this paper, we explore the pivotal subject of leveraging the generative capacity and efficiency of consistency models to facilitate controllable visual content creation via ControlNet. First, it is observed that ControlNet trained for diffusion models (DMs) can be directly applied to CMs for high-level semantic controls but sacrifice image low-level details and realism. To tackle with this issue, we develop a CMs-tailored training strategy for ControlNet using the consistency training. It is substantiated that ControlNet can be successfully established through the consistency training technique. Besides, a unified adapter can be trained utilizing the consistency training, which enhances the adaptation of DM's ControlNet. We quantitatively and qualitatively evaluate all strategies across various conditional controls, including sketch, hed, canny, depth, human pose, low-resolution image and masked image, with the pre-trained text-to-image latent consistency models.

Deep Learning · Large Language Models

Shengqiong Wu, Hao Fei, Leigang Qu, Wei Ji, Tat-Seng Chua

While recently Multimodal Large Language Models (MM-LLMs) have made exciting strides, they mostly fall prey to the limitation of only input-side multimodal understanding, without the ability to produce content in multiple modalities. As we humans always perceive the world and communicate with people through various modalities, developing any-to-any MM-LLMs capable of accepting and delivering content in any modality becomes essential to human-level AI. To fill the gap, we present an end-to-end general-purpose any-to-any MM-LLM system, NExT-GPT. We connect an LLM with multimodal adaptors and different diffusion decoders, enabling NExT-GPT to perceive inputs and generate outputs in arbitrary combinations of text, image, video, and audio. By leveraging the existing well-trained high-performing encoders and decoders, NExT-GPT is tuned with only a small amount of parameter (1%) of certain projection layers, which not only benefits low-cost training but also facilitates convenient expansion to more potential modalities. Moreover, we introduce a modality-switching instruction tuning (MosIT) and manually curate a high-quality dataset for MosIT, based on which NExT-GPT is empowered with complex cross-modal semantic understanding and content generation. Overall, our research showcases the promising possibility of building a unified AI agent capable of modeling universal modalities, paving the way for more human-like AI research in the community.

Miscellaneous Aspects of Machine Learning · Unsupervised and Semi-supervised Learning

Jie Wen, Shijie Deng, Waikeung Wong, Guoqing Chao, Chao Huang, Lunke Fei, Yong Xu

As a branch of clustering, multi-view clustering has received much attention in recent years. In practical applications, a common phenomenon is that partial views of some samples may be missing in the collected multi-view data, which poses a severe challenge to design the multi-view learning model and explore complementary and consistent information. Currently, most of the incomplete multi-view clustering methods only focus on exploring the information of available views while few works study the missing view recovery for incomplete multi-view learning. To this end, we propose an innovative diffusion-based missing view generation (DMVG) network. Moreover, for the scenarios with high missing rates, we further propose an incomplete multi-view data augmentation strategy to enhance the recovery quality for the missing views. Extensive experimental results show that the proposed DMVG can not only accurately predict missing views, but also further enhance the subsequent clustering performance in comparison with several state-of-the-art incomplete multi-view clustering methods.

Deep Learning · Generative Models and Autoencoders

Jiawei Wang, Yuchen Zhang, Jiaxin Zou, Yan Zeng, Guoqiang Wei, Liping Yuan, Hang Li

Generating rich and controllable motion is a pivotal challenge in video synthesis. We propose *Boximator*, a new approach for fine-grained motion control. Boximator introduces two constraint types: *hard box* and *soft box*. Users select objects in the conditional frame using hard boxes and then use either type of boxes to roughly or rigorously define the object’s position, shape, or motion path in future frames. Boximator functions as a plug-in for existing video diffusion models. Its training process preserves the base model’s knowledge by freezing the original weights and training only the control module. To address training challenges, we introduce a novel *self-tracking* technique that greatly simplifies the learning of box-object correlations. Empirically, Boximator achieves state-of-the-art video quality (FVD) scores, improving on two base models, and further enhanced after incorporating box constraints. Its robust motion controllability is validated by drastic increases in the bounding box alignment metric. Human evaluation also shows that users favor Boximator generation results over the base model.

Yi Wang, Zhiren Wang

We examine both the manifold hypothesis (Bengio et al., 2013) and the union of manifold hypothesis (Brown et al., 2023), and argue that, in contrast to these hypotheses, the local intrinsic dimension varies from point to point even in the same connected component. We propose an alternative CW complex hypothesis that image data is distributed in ``manifolds with skeletons". We support the hypothesis through visualization of distributions of image data of random geometric objects, as well as by introducing and testing a criterion on natural image datasets. One motivation of our work is to explain why diffusion models have difficulty generating accurate higher dimensional details such as human hands. Under the CW complex hypothesis and with both theoretical and empirical evidences, we provide an interpretation that the mixture of higher and lower dimensional components in data obstructs diffusion models from efficient learning.

Zhenting Wang, Vikash Sehwag, Chen Chen, Lingjuan Lyu, Dimitris Metaxas, Shiqing Ma

Latent generative models (e.g., Stable Diffusion) have become more and more popular, but concerns have arisen regarding potential misuse related to images generated by these models. It is, therefore, necessary to analyze the origin of images by inferring if a particular image was generated by a specific latent generative model. Most existing methods (e.g., image watermark and model fingerprinting) require extra steps during training or generation. These requirements restrict their usage on the generated images without such extra operations, and the extra required operations might compromise the quality of the generated images. In this work, we ask whether it is possible to effectively and efficiently trace the images generated by a specific latent generative model without the aforementioned requirements. To study this problem, we design a latent inversion based method called LatentTracer to trace the generated images of the inspected model by checking if the examined images can be well-reconstructed with an inverted latent input. We leverage gradient based latent inversion and identify a encoder-based initialization critical to the success of our approach. Our experiments on the state-of-the-art latent generative models, such as Stable Diffusion, show that our method can distinguish the images generated by the inspected model and other images with a high accuracy and efficiency. Our findings suggest the intriguing possibility that today's latent generative generated images are naturally watermarked by the decoder used in the source models. Code: https://github.com/ZhentingWang/LatentTracer.

Deep Learning · Generative Models and Autoencoders

chentong wang, Yannan Qu, Zhangzhi Peng, Yukai Wang, Hongli Zhu, dachuan chen, Longxing Cao

Diffusion-based generative models have been successfully employed to create proteins with novel structures and functions. However, the construction of such models typically depends on large, pre-trained structure prediction networks, like RFdiffusion. In contrast, alternative models that are trained from scratch, such as FrameDiff, still fall short in performance. In this context, we introduce Proteus, an innovative deep diffusion network that incorporates graph-based triangle methods and a multi-track interaction network, eliminating the dependency on structure prediction pre-training with superior efficiency. We have validated our model's performance on de novo protein backbone generation through comprehensive in silico evaluations and experimental characterizations, which demonstrate a remarkable success rate. These promising results underscore Proteus's ability to generate highly designable protein backbones efficiently. This capability, achieved without reliance on pre-training techniques, has the potential to significantly advance the field of protein design.