Author name cluster

Songhua Liu

Possible papers associated with this exact author name in Arrow. This page groups case-insensitive exact name matches and is not a full identity disambiguation profile.

12 papers

2 author rows

EAAI Journal 2026 Journal Article

A consistency-driven pseudo-labeling framework for robust functional connectivity modeling in neuropsychiatric disorder diagnosis

Xin Wen
Shijie Guo
Li Dong
Xiaobo Liu
Wenbo Ning
Jie Shi
Songhua Liu
Cheng Luo

The incidence of neuropsychiatric disorders such as Autism Spectrum Disorder (ASD), Attention Deficit Hyperactivity Disorder (ADHD), and Major Depressive Disorder (MDD) continues to rise. Deep learning-based computer-aided diagnosis (CAD) has emerged as a promising approach to alleviate the increasing burden on neuroimaging-based clinical resources. However, neuroimaging modalities such as functional magnetic resonance imaging (fMRI) involve complex spatiotemporal characteristics, making their representations susceptible to various types of noise and interference, which in turn hampers the effectiveness of CAD. To address this challenge, we propose a pseudo-label consistency-driven framework for functional connectivity (FC) reconstruction and discriminative modeling (PL-FCDM), aiming to enhance both the representational quality and discriminative power of FC features. Specifically, two complementary pseudo-labeling models are developed to independently capture discriminative features from the temporal domain (time series) and spatial domain (dynamic functional connectivity), enabling pseudo label prediction from distinct modalities. Then a consistency-based filtering strategy is applied to construct high-confidence reconstructed functional connectivity. These graphs are subsequently fed into a classification model comprising a Feature Optimization Autoencoder and a Depthwise Separable Convolutional Neural Network for efficient identification of neuropsychiatric disorders. Extensive experiments conducted on four publicly available multi-site datasets—ABIDE I, ABIDE II, ADHD-200, and REST-meta-MDD demonstrate that the proposed method achieves classification accuracies of 76. 14%, 74. 37%, 72. 89%, and 71. 15%, respectively. These results consistently outperform several state-of-the-art approaches, validating the effectiveness and robustness of the proposed framework in feature refinement and multi-disorder recognition.

Details DOI

AAAI Conference 2026 Conference Paper

FreLay: Frequency-aware Energy Function for Training-free Layout-to-Image Generation

Bonan Li
Yinhan Hu
Songhua Liu
Zeyu Xiao
Xinchao Wang

Layout-to-Image generation has significantly advanced content creation by enabling the rendering of visual text under predefined spatial layouts. Current approaches achieve training-free layout guidance by constructing attention-based energy functions to derive correction gradients. In this paper, we demonstrate that vanilla energy functions suffer from two limitations, resulting in imprecise layout control and visually unrealistic artifacts. First, the normalizing factor of the Boltzmann distribution defined by the energy functions is non-negligible when calculating correction gradients, yet current energy functions cannot compute this factor exactly. Furthermore, while attention varies over time during the denoising process, existing approaches employ a fixed formulation. To address these challenges, we introduce FreLay, a novel training-free approach equipped with a frequency-aware energy function. Our method first reformulates the energy function to handle the normalization factor, enabling accurate computation of correction gradients. Simultaneously, leveraging the prior knowledge that low-frequency information deteriorates slower during noise addition, we design a time-specific energy function for each timestep from a frequency-domain perspective. Experimental results demonstrate that FreLay consistently outperforms existing state-of-the-art training-free methods by a large margin both qualitatively and quantitatively across multiple datasets.

PDF Details DOI

NeurIPS Conference 2025 Conference Paper

CLEAR: Conv-Like Linearization Revs Pre-Trained Diffusion Transformers Up

Songhua Liu
Zhenxiong Tan
Xinchao Wang

Diffusion Transformers (DiT) have become a leading architecture in image generation. However, the quadratic complexity of attention mechanisms, which are responsible for modeling token-wise relationships, results in significant latency when generating high-resolution images. To address this issue, we aim at a linear attention mechanism in this paper that reduces the complexity of pre-trained DiTs to linear. We begin our exploration with a comprehensive summary of existing efficient attention mechanisms and identify four key factors crucial for successful linearization of pre-trained DiTs: locality, formulation consistency, high-rank attention maps, and feature integrity. Based on these insights, we introduce a convolution-like local attention strategy termed CLEAR, which limits feature interactions to a local window around each query token, and thus achieves linear complexity. Our experiments indicate that, by fine-tuning the attention layer on merely 10K self-generated samples for 10K iterations, we can effectively transfer knowledge from a pre-trained DiT to a student model with linear complexity, yielding results comparable to the teacher model. Simultaneously, it reduces attention computations by 99. 5% and accelerates generation by 6. 3 times for generating 8K-resolution images. Furthermore, we investigate favorable properties in the distilled attention layers, such as zero-shot generalization across various models and plugins, and improved support for multi-GPU parallel inference. Models and codes will be available.

PDF Details

ICML Conference 2025 Conference Paper

Control and Realism: Best of Both Worlds in Layout-to-Image without Training

Bonan Li
Yinhan Hu
Songhua Liu
Xinchao Wang

Layout-to-Image generation aims to create complex scenes with precise control over the placement and arrangement of subjects. Existing works have demonstrated that pre-trained Text-to-Image diffusion models can achieve this goal without training on any specific data; however, they often face challenges with imprecise localization and unrealistic artifacts. Focusing on these drawbacks, we propose a novel training-free method, WinWinLay. At its core, WinWinLay presents two key strategies—Non-local Attention Energy Function and Adaptive Update—that collaboratively enhance control precision and realism. On one hand, we theoretically demonstrate that the commonly used attention energy function introduces inherent spatial distribution biases, hindering objects from being uniformly aligned with layout instructions. To overcome this issue, non-local attention prior is explored to redistribute attention scores, facilitating objects to better conform to the specified spatial conditions. On the other hand, we identify that the vanilla backpropagation update rule can cause deviations from the pre-trained domain, leading to out-of-distribution artifacts. We accordingly introduce a Langevin dynamics-based adaptive update scheme as a remedy that promotes in-domain updating while respecting layout constraints. Extensive experiments demonstrate that WinWinLay excels in controlling element placement and achieving photorealistic visual fidelity, outperforming the current state-of-the-art methods.

Details

NeurIPS Conference 2025 Conference Paper

Image Editing As Programs with Diffusion Models

Yujia Hu
Songhua Liu
Zhenxiong Tan
Xingyi Yang
Xinchao Wang

While diffusion models have achieved remarkable success in text-to-image generation, they encounter significant challenges with instruction-driven image editing. Our research highlights a key challenge: these models particularly struggle with structurally-inconsistent edits that involve substantial layout changes. To address this gap, we introduce Image Editing As Programs (IEAP), a unified image editing framework built upon the Diffusion Transformer (DiT) architecture. Specifically, IEAP deals with complex instructions by decomposing them into a sequence of programmable atomic operations. Each atomic operation manages a specific type of structurally consistent edit; when sequentially combined, IEAP enables the execution of arbitrary and structurally-inconsistent transformations. This reductionist approach enables IEAP to robustly handle a wide spectrum of edits, encompassing both structurally-consistent and inconsistent changes. Extensive experiments demonstrate that IEAP significantly outperforms state-of-the-art methods on standard benchmarks across various editing scenarios. In these evaluations, our framework delivers superior accuracy and semantic fidelity, particularly for complex, multi-step instructions. Codes are available at https: //github. com/YujiaHu1109/IEAP.

PDF Details

ICML Conference 2025 Conference Paper

Ultra-Resolution Adaptation with Ease

Ruonan Yu
Songhua Liu
Zhenxiong Tan
Xinchao Wang

Text-to-image diffusion models have achieved remarkable progress in recent years. However, training models for high-resolution image generation remains challenging, particularly when training data and computational resources are limited. In this paper, we explore this practical problem from two key perspectives: data and parameter efficiency, and propose a set of key guidelines for ultra-resolution adaptation termed URAE. For data efficiency, we theoretically and empirically demonstrate that synthetic data generated by some teacher models can significantly promote training convergence. For parameter efficiency, we find that tuning minor components of the weight matrices outperforms widely-used low-rank adapters when synthetic data are unavailable, offering substantial performance gains while maintaining efficiency. Additionally, for models leveraging guidance distillation, such as FLUX, we show that disabling classifier-free guidance, i. e. , setting the guidance scale to 1 during adaptation, is crucial for satisfactory performance. Extensive experiments validate that URAE achieves comparable 2K-generation performance to state-of-the-art closed-source models like FLUX1. 1 [Pro] Ultra with only 3K samples and 2K iterations, while setting new benchmarks for 4K-resolution generation. Codes are available here.

Details

AAAI Conference 2024 Conference Paper

Mutual-Modality Adversarial Attack with Semantic Perturbation

Jingwen Ye
Ruonan Yu
Songhua Liu
Xinchao Wang

Adversarial attacks constitute a notable threat to machine learning systems, given their potential to induce erroneous predictions and classifications. However, within real-world contexts, the essential specifics of the deployed model are frequently treated as a black box, consequently mitigating the vulnerability to such attacks. Thus, enhancing the transferability of the adversarial samples has become a crucial area of research, which heavily relies on selecting appropriate surrogate models. To address this challenge, we propose a novel approach that generates adversarial attacks in a mutual-modality optimization scheme. Our approach is accomplished by leveraging the pre-trained CLIP model. Firstly, we conduct a visual attack on the clean image that causes semantic perturbations on the aligned embedding space with the other textual modality. Then, we apply the corresponding defense on the textual modality by updating the prompts, which forces the re-matching on the perturbed embedding space. Finally, to enhance the attack transferability, we utilize the iterative training strategy on the visual attack and the textual defense, where the two processes optimize from each other. We evaluate our approach on several benchmark datasets and demonstrate that our mutual-modal attack strategy can effectively produce high-transferable attacks, which are stable regardless of the target networks. Our approach outperforms state-of-the-art attack methods and can be readily deployed as a plug-and-play solution.

PDF Details DOI

NeurIPS Conference 2024 Conference Paper

One-shot Federated Learning via Synthetic Distiller-Distillate Communication

Junyuan Zhang
Songhua Liu
Xinchao Wang

One-shot Federated learning (FL) is a powerful technology facilitating collaborative training of machine learning models in a single round of communication. While its superiority lies in communication efficiency and privacy preservation compared to iterative FL, one-shot FL often compromises model performance. Prior research has primarily focused on employing data-free knowledge distillation to optimize data generators and ensemble models for better aggregating local knowledge into the server model. However, these methods typically struggle with data heterogeneity, where inconsistent local data distributions can cause teachers to provide misleading knowledge. Additionally, they may encounter scalability issues with complex datasets due to inherent two-step information loss: first, during local training (from data to model), and second, when transferring knowledge to the server model (from model to inversed data). In this paper, we propose FedSD2C, a novel and practical one-shot FL framework designed to address these challenges. FedSD2C introduces a distiller to synthesize informative distillates directly from local data to reduce information loss and proposes sharing synthetic distillates instead of inconsistent local models to tackle data heterogeneity. Our empirical results demonstrate that FedSD2C consistently outperforms other one-shot FL methods with more complex and real datasets, achieving up to 2. 6 $\times$ the performance of the best baseline. Code: https: //github. com/Carkham/FedSD2C

PDF Details DOI

ICML Conference 2024 Conference Paper

StyDeSty: Min-Max Stylization and Destylization for Single Domain Generalization

Songhua Liu
Xin Jin
Xingyi Yang
Jingwen Ye
Xinchao Wang

Single domain generalization (single DG) aims at learning a robust model generalizable to unseen domains from only one training domain, making it a highly ambitious and challenging task. State-of-the-art approaches have mostly relied on data augmentations, such as adversarial perturbation and style enhancement, to synthesize new data and thus increase robustness. Nevertheless, they have largely overlooked the underlying coherence between the augmented domains, which in turn leads to inferior results in real-world scenarios. In this paper, we propose a simple yet effective scheme, termed as StyDeSty, to explicitly account for the alignment of the source and pseudo domains in the process of data augmentation, enabling them to interact with each other in a self-consistent manner and further giving rise to a latent domain with strong generalization power. The heart of StyDeSty lies in the interaction between a stylization module for generating novel stylized samples using the source domain, and a destylization module for transferring stylized and source samples to a latent domain to learn content-invariant features. The stylization and destylization modules work adversarially and reinforce each other. During inference, the destylization module transforms the input sample with an arbitrary style shift to the latent domain, in which the downstream tasks are carried out. Specifically, the location of the destylization layer within the backbone network is determined by a dedicated neural architecture search (NAS) strategy. We evaluate StyDeSty on multiple benchmarks and demonstrate that it yields encouraging results, outperforming the state of the art by up to 13. 44% on classification accuracy. Codes are available https: //github. com/Huage001/StyDeSty.

Details

NeurIPS Conference 2023 Conference Paper

MGDD: A Meta Generator for Fast Dataset Distillation

Songhua Liu
Xinchao Wang

Existing dataset distillation (DD) techniques typically rely on iterative strategies to synthesize condensed datasets, where datasets before and after distillation are forward and backward through neural networks a massive number of times. Despite the promising results achieved, the time efficiency of prior approaches is still far from satisfactory. Moreover, when different sizes of synthetic datasets are required, they have to repeat the iterative training procedures, which is highly cumbersome and lacks flexibility. In this paper, different from the time-consuming forward-backward passes, we introduce a generative fashion for dataset distillation with significantly improved efficiency. Specifically, synthetic samples are produced by a generator network conditioned on the initialization of DD, while synthetic labels are obtained by solving a least-squares problem in a feature space. Our theoretical analysis reveals that the errors of synthetic datasets solved in the original space and then processed by any conditional generators are upper-bounded. To find a satisfactory generator efficiently, we propose a meta-learning algorithm, where a meta generator is trained on a large dataset so that only a few steps are required to adapt to a target dataset. The meta generator is termed as MGDD in our approach. Once adapted, it can handle arbitrary sizes of synthetic datasets, even for those unseen during adaptation. Experiments demonstrate that the generator adapted with only a limited number of steps performs on par with those state-of-the-art DD methods and yields $22\times$ acceleration.

PDF Details

NeurIPS Conference 2022 Conference Paper

Dataset Distillation via Factorization

Songhua Liu
Kai Wang
Xingyi Yang
Jingwen Ye
Xinchao Wang

In this paper, we study dataset distillation (DD), from a novel perspective and introduce a \emph{dataset factorization} approach, termed \emph{HaBa}, which is a plug-and-play strategy portable to any existing DD baseline. Unlike conventional DD approaches that aim to produce distilled and representative samples, \emph{HaBa} explores decomposing a dataset into two components: data \emph{Ha}llucination networks and \emph{Ba}ses, where the latter is fed into the former to reconstruct image samples. The flexible combinations between bases and hallucination networks, therefore, equip the distilled data with exponential informativeness gain, which largely increase the representation capability of distilled datasets. To furthermore increase the data efficiency of compression results, we further introduce a pair of adversarial contrastive \xw{constraints} on the resultant hallucination networks and bases, which increase the diversity of generated images and inject more discriminant information into the factorization. Extensive comparisons and experiments demonstrate that our method can yield significant improvement on downstream classification tasks compared with previous state of the arts, while reducing the total number of compressed parameters by up to 65\%. Moreover, distilled datasets by our approach also achieve \textasciitilde10\% higher accuracy than baseline methods in cross-architecture generalization. Our code is available \href{https: //github. com/Huage001/DatasetFactorization}{here}.

PDF Details

NeurIPS Conference 2022 Conference Paper

Deep Model Reassembly

Xingyi Yang
Daquan Zhou
Songhua Liu
Jingwen Ye
Xinchao Wang

In this paper, we explore a novel knowledge-transfer task, termed as Deep Model Reassembly (DeRy), for general-purpose model reuse. Given a collection of heterogeneous models pre-trained from distinct sources and with diverse architectures, the goal of DeRy, as its name implies, is to first dissect each model into distinctive building blocks, and then selectively reassemble the derived blocks to produce customized networks under both the hardware resource and performance constraints. Such ambitious nature of DeRy inevitably imposes significant challenges, including, in the first place, the feasibility of its solution. We strive to showcase that, through a dedicated paradigm proposed in this paper, DeRy can be made not only possibly but practically efficiently. Specifically, we conduct the partitions of all pre-trained networks jointly via a cover set optimization, and derive a number of equivalence set, within each of which the network blocks are treated as functionally equivalent and hence interchangeable. The equivalence sets learned in this way, in turn, enable picking and assembling blocks to customize networks subject to certain constraints, which is achieved via solving an integer program backed up with a training-free proxy to estimate the task performance. The reassembled models give rise to gratifying performances with the user-specified constraints satisfied. We demonstrate that on ImageNet, the best reassemble model achieves 78. 6% top-1 accuracy without fine-tuning, which could be further elevated to 83. 2% with end-to-end fine-tuning. Our code is available at https: //github. com/Adamdad/DeRy.

PDF Details