NeurIPS 2024 论文列表

We introduce Equivariant Neural Diffusion (END), a novel diffusion model for molecule generation in 3D that is equivariant to Euclidean transformations. Compared to current state-of-the-art equivariant diffusion models, the key innovation in END lies in its learnable forward process for enhanced generative modelling. Rather than pre-specified, the forward process is parameterized through a time- and data-dependent transformation that is equivariant to rigid transformations. Through a series of experiments on standard molecule generation benchmarks, we demonstrate the competitive performance of END compared to several strong baselines for both unconditional and conditional generation.

Generating realistic images from arbitrary views based on a single source image remains a significant challenge in computer vision, with broad applications ranging from e-commerce to immersive virtual experiences. Recent advancements in diffusion models, particularly the Zero-1-to-3 model, have been widely adopted for generating plausible views, videos, and 3D models. However, these models still struggle with inconsistencies and implausibility in new views generation, especially for challenging changes in viewpoint. In this work, we propose Zero-to-Hero, a novel test-time approach that enhances view synthesis by manipulating attention maps during the denoising process of Zero-1-to-3. By drawing an analogy between the denoising process and stochastic gradient descent (SGD), we implement a filtering mechanism that aggregates attention maps, enhancing generation reliability and authenticity. This process improves geometric consistency without requiring retraining or significant computational resources. Additionally, we modify the self-attention mechanism to integrate information from the source view, reducing shape distortions. These processes are further supported by a specialized sampling schedule. Experimental results demonstrate substantial improvements in fidelity and consistency, validated on a diverse set of out-of-distribution objects. Additionally, we demonstrate the general applicability and effectiveness of Zero-to-Hero in multi-view, and image generation conditioned on semantic maps and pose.

Offline safe reinforcement learning (RL) aims to train a policy that satisfies con- straints using a pre-collected dataset. Most current methods struggle with the mismatch between imperfect demonstrations and the desired safe and rewarding performance. In this paper, we mitigate this issue from a data-centric perspective and introduce OASIS (cOnditionAl diStributIon Shaping), a new paradigm in offline safe RL designed to overcome these critical limitations. OASIS utilizes a conditional diffusion model to synthesize offline datasets, thus shaping the data dis- tribution toward a beneficial target domain. Our approach makes compliance with safety constraints through effective data utilization and regularization techniques to benefit offline safe RL training. Comprehensive evaluations on public benchmarks and varying datasets showcase OASIS’s superiority in benefiting offline safe RL agents to achieve high-reward behavior while satisfying the safety constraints, out- performing established baselines. Furthermore, OASIS exhibits high data efficiency and robustness, making it suitable for real-world applications, particularly in tasks where safety is imperative and high-quality demonstrations are scarce. More details are available at the website https://sites.google.com/view/saferl-oasis/home.

Unsupervised out-of-distribution (OOD) detection aims to identify out-of-domain data by learning only from unlabeled In-Distribution (ID) training samples, which is crucial for developing a safe real-world machine learning system. Current reconstruction-based method provides a good alternative approach, by measuring the reconstruction error between the input and its corresponding generative counterpart in the pixel/feature space. However, such generative methods face the key dilemma, $i.e.$, improving the reconstruction power of the generative model, while keeping compact representation of the ID data. To address this issue, we propose the diffusion-based layer-wise semantic reconstruction approach for unsupervised OOD detection. The innovation of our approach is that we leverage the diffusion model's intrinsic data reconstruction ability to distinguish ID samples from OOD samples in the latent feature space. Moreover, to set up a comprehensive and discriminative feature representation, we devise a multi-layer semantic feature extraction strategy. Through distorting the extracted features with Gaussian noises and applying the diffusion model for feature reconstruction, the separation of ID and OOD samples is implemented according to the reconstruction errors. Extensive experimental results on multiple benchmarks built upon various datasets demonstrate that our method achieves state-of-the-art performance in terms of detection accuracy and speed.

Recent advancements in text-to-video generative models, such as Sora, have showcased impressive capabilities. These models have attracted significant interest for their potential applications. However, they often rely on extensive datasets of variable quality, which can result in generated videos that lack aesthetic appeal and do not accurately reflect the input text prompts. A promising approach to mitigate these issues is to leverage Reinforcement Learning from Human Feedback (RLHF), which aims to align the outputs of text-to-video generative with human preferences. However, the considerable costs associated with manual annotation have led to a scarcity of comprehensive preference datasets. In response to this challenge, our study begins by investigating the efficacy of Multimodal Large Language Models (MLLMs) generated annotations in capturing video preferences, discovering a high degree of concordance with human judgments. Building upon this finding, we utilize MLLMs to perform fine-grained video preference annotations across two dimensions, resulting in the creation of VideoPrefer, which includes 135,000 preference annotations. Utilizing this dataset, we introduce VideoRM, the first general-purpose reward model tailored for video preference in the text-to-video domain. Our comprehensive experiments confirm the effectiveness of both VideoPrefer and VideoRM, representing a significant step forward in the field.

Event cameras harness advantages such as low latency, high temporal resolution, and high dynamic range (HDR), compared to standard cameras. Due to the distinct imaging paradigm shift, a dominant line of research focuses on event-to-video (E2V) reconstruction to bridge event-based and standard computer vision. However, this task remains challenging due to its inherently ill-posed nature: event cameras only detect the edge and motion information locally. Consequently, the reconstructed videos are often plagued by artifacts and regional blur, primarily caused by the ambiguous semantics of event data. In this paper, we find language naturally conveys abundant semantic information, rendering it stunningly superior in ensuring semantic consistency for E2V reconstruction. Accordingly, we propose a novel framework, called LaSe-E2V, that can achieve semantic-aware high-quality E2V reconstruction from a language-guided perspective, buttressed by the text-conditional diffusion models. However, due to diffusion models' inherent diversity and randomness, it is hardly possible to directly apply them to achieve spatial and temporal consistency for E2V reconstruction. Thus, we first propose an Event-guided Spatiotemporal Attention (ESA) module to condition the event data to the denoising pipeline effectively. We then introduce an event-aware mask loss to ensure temporal coherence and a noise initialization strategy to enhance spatial consistency. Given the absence of event-text-video paired data, we aggregate existing E2V datasets and generate textual descriptions using the tagging models for training and evaluation. Extensive experiments on three datasets covering diverse challenging scenarios (e.g., fast motion, low light) demonstrate the superiority of our method. Demo videos for the results are attached to the project page.

Efficient computation of graph diffusion equations (GDEs), such as Personalized PageRank, Katz centrality, and the Heat kernel, is crucial for clustering, training neural networks, and many other graph-related problems. Standard iterative methods require accessing the whole graph per iteration, making them time-consuming for large-scale graphs. While existing local solvers approximate diffusion vectors through heuristic local updates, they often operate sequentially and are typically designed for specific diffusion types, limiting their applicability. Given that diffusion vectors are highly localizable, as measured by the participation ratio, this paper introduces a novel framework for approximately solving GDEs using a local diffusion process. This framework reveals the suboptimality of existing local solvers. Furthermore, our approach effectively localizes standard iterative solvers by designing simple and provably sublinear time algorithms. These new local solvers are highly parallelizable, making them well-suited for implementation on GPUs. We demonstrate the effectiveness of our framework in quickly obtaining approximate diffusion vectors, achieving up to a hundred-fold speed improvement, and its applicability to large-scale dynamic graphs. Our framework could also facilitate more efficient local message-passing mechanisms for GNNs.

Complex 3D scene understanding has gained increasing attention, with scene encoding strategies built on top of visual foundation models playing a crucial role in this success. However, the optimal scene encoding strategies for various scenarios remain unclear, particularly compared to their image-based counterparts. To address this issue, we present the first comprehensive study that probes various visual encoding models for 3D scene understanding, identifying the strengths and limitations of each model across different scenarios. Our evaluation spans seven vision foundation encoders, including image, video, and 3D foundation models. We evaluate these models in four tasks: Vision-Language Scene Reasoning, Visual Grounding, Segmentation, and Registration, each focusing on different aspects of scene understanding. Our evaluation yields key intriguing findings: Unsupervised image foundation models demonstrate superior overall performance, video models excel in object-level tasks, diffusion models benefit geometric tasks, language-pretrained models show unexpected limitations in language-related tasks, and the mixture-of-vision-expert (MoVE) strategy leads to consistent performance improvement. These insights challenge some conventional understandings, provide novel perspectives on leveraging visual foundation models, and highlight the need for more flexible encoder selection in future vision-language and scene understanding tasks.

Beyond scaling base models with more data or parameters, fine-tuned adapters provide an alternative way to generate high fidelity, custom images at reduced costs. As such, adapters have been widely adopted by open-source communities, accumulating a database of over 100K adapters—most of which are highly customized with insufficient descriptions. To generate high quality images, this paper explores the problem of matching the prompt to a Stylus of relevant adapters, built on recent work that highlight the performance gains of composing adapters. We introduce Stylus, which efficiently selects and automatically composes task-specific adapters based on a prompt's keywords. Stylus outlines a three-stage approach that first summarizes adapters with improved descriptions and embeddings, retrieves relevant adapters, and then further assembles adapters based on prompts' keywords by checking how well they fit the prompt. To evaluate Stylus, we developed StylusDocs, a curated dataset featuring 75K adapters with pre-computed adapter embeddings. In our evaluation on popular Stable Diffusion checkpoints, Stylus achieves greater CLIP/FID Pareto efficiency and is twice as preferred, with humans and multimodal models as evaluators, over the base model.

This paper presents innovative enhancements to diffusion models by integrating a novel multi-resolution network and time-dependent layer normalization. Diffusion models have gained prominence for their effectiveness in high-fidelity image generation. While conventional approaches rely on convolutional U-Net architectures, recent Transformer-based designs have demonstrated superior performance and scalability. However, Transformer architectures, which tokenize input data (via "patchification"), face a trade-off between visual fidelity and computational complexity due to the quadratic nature of self-attention operations concerning token length. While larger patch sizes enable attention computation efficiency, they struggle to capture fine-grained visual details, leading to image distortions. To address this challenge, we propose augmenting the **Di**ffusion model with the **M**ulti-**R**esolution network (DiMR), a framework that refines features across multiple resolutions, progressively enhancing detail from low to high resolution. Additionally, we introduce Time-Dependent Layer Normalization (TD-LN), a parameter-efficient approach that incorporates time-dependent parameters into layer normalization to inject time information and achieve superior performance. Our method's efficacy is demonstrated on the class-conditional ImageNet generation benchmark, where DiMR-XL variants surpass previous diffusion models, achieving FID scores of 1.70 on ImageNet $256 \times 256$ and 2.89 on ImageNet $512 \times 512$. Our best variant, DiMR-G, further establishes a state-of-the-art 1.63 FID on ImageNet $256 \times 256$.

The Diffusion Model has not only garnered noteworthy achievements in the realm of image generation but has also demonstrated its potential as an effective pretraining method utilizing unlabeled data. Drawing from the extensive potential unveiled by the Diffusion Model in both semantic correspondence and open vocabulary segmentation, our work initiates an investigation into employing the Latent Diffusion Model for Few-shot Semantic Segmentation. Recently, inspired by the in-context learning ability of large language models, Few-shot Semantic Segmentation has evolved into In-context Segmentation tasks, morphing into a crucial element in assessing generalist segmentation models. In this context, we concentrate on Few-shot Semantic Segmentation, establishing a solid foundation for the future development of a Diffusion-based generalist model for segmentation. Our initial focus lies in understanding how to facilitate interaction between the query image and the support image, resulting in the proposal of a KV fusion method within the self-attention framework. Subsequently, we delve deeper into optimizing the infusion of information from the support mask and simultaneously re-evaluating how to provide reasonable supervision from the query mask. Based on our analysis, we establish a simple and effective framework named DiffewS, maximally retaining the original Latent Diffusion Model's generative framework and effectively utilizing the pre-training prior. Experimental results demonstrate that our method significantly outperforms the previous SOTA models in multiple settings.

We introduce X-Ray, a novel 3D sequential representation inspired by the penetrability of x-ray scans. X-Ray transforms a 3D object into a series of surface frames at different layers, making it suitable for generating 3D models from images. Our method utilizes ray casting from the camera center to capture geometric and textured details, including depth, normal, and color, across all intersected surfaces. This process efficiently condenses the whole 3D object into a multi-frame video format, motivating the utilize of a network architecture similar to those in video diffusion models. This design ensures an efficient 3D representation by focusing solely on surface information. Also, we propose a two-stage pipeline to generate 3D objects from X-Ray Diffusion Model and Upsampler. We demonstrate the practicality and adaptability of our X-Ray representation by synthesizing the complete visible and hidden surfaces of a 3D object from a single input image. Experimental results reveal the state-of-the-art superiority of our representation in enhancing the accuracy of 3D generation, paving the way for new 3D representation research and practical applications. Our project page is in \url{https://tau-yihouxiang.github.io/projects/X-Ray/X-Ray.html}.

By training over large-scale datasets, zero-shot monocular depth estimation (MDE) methods show robust performance in the wild but often suffer from insufficient detail. Although recent diffusion-based MDE approaches exhibit a superior ability to extract details, they struggle in geometrically complex scenes that challenge their geometry prior, trained on less diverse 3D data. To leverage the complementary merits of both worlds, we propose BetterDepth to achieve geometrically correct affine-invariant MDE while capturing fine details. Specifically, BetterDepth is a conditional diffusion-based refiner that takes the prediction from pre-trained MDE models as depth conditioning, in which the global depth layout is well-captured, and iteratively refines details based on the input image. For the training of such a refiner, we propose global pre-alignment and local patch masking methods to ensure BetterDepth remains faithful to the depth conditioning while learning to add fine-grained scene details. With efficient training on small-scale synthetic datasets, BetterDepth achieves state-of-the-art zero-shot MDE performance on diverse public datasets and on in-the-wild scenes. Moreover, BetterDepth can improve the performance of other MDE models in a plug-and-play manner without further re-training.

In image editing, Denoising Diffusion Implicit Models (DDIM) inversion has become a widely adopted method and is extensively used in various image editing approaches. The core concept of DDIM inversion stems from the deterministic sampling technique of DDIM, which allows the DDIM process to be viewed as an Ordinary Differential Equation (ODE) process that is reversible. This enables the prediction of corresponding noise from a reference image, ensuring that the restored image from this noise remains consistent with the reference image. Image editing exploits this property by modifying the cross-attention between text and images to edit specific objects while preserving the remaining regions. However, in the DDIM inversion, using the $t-1$ time step to approximate the noise prediction at time step $t$ introduces errors between the restored image and the reference image. Recent approaches have modeled each step of the DDIM inversion process as finding a fixed-point problem of an implicit function. This approach significantly mitigates the error in the restored image but lacks theoretical support regarding the existence of such fixed points. Therefore, this paper focuses on the study of fixed points in DDIM inversion and provides theoretical support. Based on the obtained theoretical insights, we further optimize the loss function for the convergence of fixed points in the original DDIM inversion, improving the visual quality of the edited image. Finally, we extend the fixed-point based image editing to the application of unsupervised image dehazing, introducing a novel text-based approach for unsupervised dehazing.

Text-to-image diffusion models benefit artists with high-quality image generation. Yet their stochastic nature hinders artists from creating consistent images of the same subject. Existing methods try to tackle this challenge and generate consistent content in various ways. However, they either depend on external restricted data or require expensive tuning of the diffusion model. For this issue, we propose a novel one-shot tuning paradigm, termed OneActor. It efficiently performs consistent subject generation solely driven by prompts via a learned semantic guidance to bypass the laborious backbone tuning. We lead the way to formalize the objective of consistent subject generation from a clustering perspective, and thus design a cluster-conditioned model. To mitigate the overfitting challenge shared by one-shot tuning pipelines, we augment the tuning with auxiliary samples and devise two inference strategies: semantic interpolation and cluster guidance. These techniques are later verified to significantly improve the generation quality. Comprehensive experiments show that our method outperforms a variety of baselines with satisfactory subject consistency, superior prompt conformity as well as high image quality. Our method is capable of multi-subject generation and compatible with popular diffusion extensions. Besides, we achieve a $4\times$ faster tuning speed than tuning-based baselines and, if desired, avoid increasing the inference time. Furthermore, our method can be naturally utilized to pre-train a consistent subject generation network from scratch, which will implement this research task into more practical applications. (Project page: https://johnneywang.github.io/OneActor-webpage/)

The recent emergence of diffusion models has significantly advanced the precision of learnable priors, presenting innovative avenues for addressing inverse problems. Previous works have endeavored to integrate diffusion priors into the maximum a posteriori estimation (MAP) framework and design optimization methods to solve the inverse problem. However, prevailing optimization-based rithms primarily exploit the prior information within the diffusion models while neglecting their denoising capability. To bridge this gap, this work leverages the diffusion process to reframe noisy inverse problems as a two-variable constrained optimization task by introducing an auxiliary optimization variable that represents a 'noisy' sample at an equivalent denoising step. The projection gradient descent method is efficiently utilized to solve the corresponding optimization problem by truncating the gradient through the $\mu$-predictor. The proposed algorithm, termed ProjDiff, effectively harnesses the prior information and the denoising capability of a pre-trained diffusion model within the optimization framework. Extensive experiments on the image restoration tasks and source separation and partial generation tasks demonstrate that ProjDiff exhibits superior performance across various linear and nonlinear inverse problems, highlighting its potential for practical applications. Code is available at https://github.com/weigerzan/ProjDiff/.

Enabling deep models to generalize in non-stationary environments is vital for real-world machine learning, as data distributions are often found to continually change. Recently, evolving domain generalization (EDG) has emerged to tackle the domain generalization in a time-varying system, where the domain gradually evolves over time in an underlying continuous structure. Nevertheless, it typically assumes multiple source domains simultaneously ready. It still remains an open problem to address EDG in the domain-incremental setting, where source domains are non-static and arrive sequentially to mimic the evolution of training domains. To this end, we propose Weight Diffusion (W-Diff), a novel framework that utilizes the conditional diffusion model in the parameter space to learn the evolving pattern of classifiers during the domain-incremental training process. Specifically, the diffusion model is conditioned on the classifier weights of different historical domain (regarded as a reference point) and the prototypes of current domain, to learn the evolution from the reference point to the classifier weights of current domain (regarded as the anchor point). In addition, a domain-shared feature encoder is learned by enforcing prediction consistency among multiple classifiers, so as to mitigate the overfitting problem and restrict the evolving pattern to be reflected in the classifier as much as possible. During inference, we adopt the ensemble manner based on a great number of target domain-customized classifiers, which are cheaply obtained via the conditional diffusion model, for robust prediction. Comprehensive experiments on both synthetic and real-world datasets show the superior generalization performance of W-Diff on unseen domains in the future.

This paper proposes a fast and general-purpose image restoration method. The key idea is to achieve few-step or even one-step inference by conducting consistency distilling or training on a specific mean-reverting stochastic differential equations. Furthermore, based on this, we propose a novel linear-nonlinear decoupling training strategy, significantly enhancing training effectiveness and surpassing consistency distillation on inference performance. This allows our method to be independent of any pre-trained checkpoint, enabling it to serve as an effective standalone image-to-image transformation model. Finally, to avoid trivial solutions and stabilize model training, we introduce a simple origin-guided loss. To validate the effectiveness of our proposed method, we conducted experiments on tasks including image deraining, denoising, deblurring, and low-light image enhancement. The experiments show that our method achieves highly competitive results with only one-step inference. And with just two-step inference, it can achieve state-of-the-art performance in low-light image enhancement. Furthermore, a number of ablation experiments demonstrate the effectiveness of the proposed training strategy. our code is available at https://github.com/XiaoxuanGong/IR-CM.

Diffusion planning has been recognized as an effective decision-making paradigm in various domains. The capability of generating high-quality long-horizon trajectories makes it a promising research direction. However, existing diffusion planning methods suffer from low decision-making frequencies due to the expensive iterative sampling cost. To alleviate this, we introduce DiffuserLite, a super fast and lightweight diffusion planning framework, which employs a planning refinement process (PRP) to generate coarse-to-fine-grained trajectories, significantly reducing the modeling of redundant information and leading to notable increases in decision-making frequency. Our experimental results demonstrate that DiffuserLite achieves a decision-making frequency of $122.2$Hz ($112.7$x faster than predominant frameworks) and reaches state-of-the-art performance on D4RL, Robomimic, and FinRL benchmarks. In addition, DiffuserLite can also serve as a flexible plugin to increase the decision-making frequency of other diffusion planning algorithms, providing a structural design reference for future works. More details and visualizations are available at https://diffuserlite.github.io/.