Yunlong Yu

Hao Chen, Yonghan Dong, Zheming Lu et al. · cmu

Few-Shot Segmentation (FSS) aims to segment the novel class images with a few annotated samples. In this paper, we propose a dense affinity matching (DAM) framework to exploit the support-query interaction by densely capturing both the pixel-to-pixel and pixel-to-patch relations in each support-query pair with the bidirectional 3D convolutions. Different from the existing methods that remove the support background, we design a hysteretic spatial filtering module (HSFM) to filter the background-related query features and retain the foreground-related query features with the assistance of the support background, which is beneficial for eliminating interference objects in the query background. We comprehensively evaluate our DAM on ten benchmarks under cross-category, cross-dataset, and cross-domain FSS tasks. Experimental results demonstrate that DAM performs very competitively under different settings with only 0.68M parameters, especially under cross-domain FSS tasks, showing its effectiveness and efficiency.

Haoran Wang, Dongliang He, Wenhao Wu et al. · amazon-science

Image-Text Retrieval (ITR) is challenging in bridging visual and lingual modalities. Contrastive learning has been adopted by most prior arts. Except for limited amount of negative image-text pairs, the capability of constrastive learning is restricted by manually weighting negative pairs as well as unawareness of external knowledge. In this paper, we propose our novel Coupled Diversity-Sensitive Momentum Constrastive Learning (CODER) for improving cross-modal representation. Firstly, a novel diversity-sensitive contrastive learning (DCL) architecture is invented. We introduce dynamic dictionaries for both modalities to enlarge the scale of image-text pairs, and diversity-sensitiveness is achieved by adaptive negative pair weighting. Furthermore, two branches are designed in CODER. One learns instance-level embeddings from image/text, and it also generates pseudo online clustering labels for its input image/text based on their embeddings. Meanwhile, the other branch learns to query from commonsense knowledge graph to form concept-level descriptors for both modalities. Afterwards, both branches leverage DCL to align the cross-modal embedding spaces while an extra pseudo clustering label prediction loss is utilized to promote concept-level representation learning for the second branch. Extensive experiments conducted on two popular benchmarks, i.e. MSCOCO and Flicker30K, validate CODER remarkably outperforms the state-of-the-art approaches.

Mushui Liu, Bozheng Li, Yunlong Yu

Recent Vision-Language Models (VLMs) \textit{e.g.} CLIP have made great progress in video recognition. Despite the improvement brought by the strong visual backbone in extracting spatial features, CLIP still falls short in capturing and integrating spatial-temporal features which is essential for video recognition. In this paper, we propose OmniCLIP, a framework that adapts CLIP for video recognition by focusing on learning comprehensive features encompassing spatial, temporal, and dynamic spatial-temporal scales, which we refer to as omni-scale features. This is achieved through the design of spatial-temporal blocks that include parallel temporal adapters (PTA), enabling efficient temporal modeling. Additionally, we introduce a self-prompt generator (SPG) module to capture dynamic object spatial features. The synergy between PTA and SPG allows OmniCLIP to discern varying spatial information across frames and assess object scales over time. We have conducted extensive experiments in supervised video recognition, few-shot video recognition, and zero-shot recognition tasks. The results demonstrate the effectiveness of our method, especially with OmniCLIP achieving a top-1 accuracy of 74.30\% on HMDB51 in a 16-shot setting, surpassing the recent MotionPrompt approach even with full training data. The code is available at \url{https://github.com/XiaoBuL/OmniCLIP}.

Huanzhang Dou, Pengyi Zhang, Wei Su et al.

Gait is one of the most promising biometrics that aims to identify pedestrians from their walking patterns. However, prevailing methods are susceptible to confounders, resulting in the networks hardly focusing on the regions that reflect effective walking patterns. To address this fundamental problem in gait recognition, we propose a Generative Counterfactual Intervention framework, dubbed GaitGCI, consisting of Counterfactual Intervention Learning (CIL) and Diversity-Constrained Dynamic Convolution (DCDC). CIL eliminates the impacts of confounders by maximizing the likelihood difference between factual/counterfactual attention while DCDC adaptively generates sample-wise factual/counterfactual attention to efficiently perceive the sample-wise properties. With matrix decomposition and diversity constraint, DCDC guarantees the model to be efficient and effective. Extensive experiments indicate that proposed GaitGCI: 1) could effectively focus on the discriminative and interpretable regions that reflect gait pattern; 2) is model-agnostic and could be plugged into existing models to improve performance with nearly no extra cost; 3) efficiently achieves state-of-the-art performance on arbitrary scenarios (in-the-lab and in-the-wild).

Huanzhang Dou, Pengyi Zhang, Wei Su et al.

Gait recognition, which aims at identifying individuals by their walking patterns, has recently drawn increasing research attention. However, gait recognition still suffers from the conflicts between the limited binary visual clues of the silhouette and numerous covariates with diverse scales, which brings challenges to the model's adaptiveness. In this paper, we address this conflict by developing a novel MetaGait that learns to learn an omni sample adaptive representation. Towards this goal, MetaGait injects meta-knowledge, which could guide the model to perceive sample-specific properties, into the calibration network of the attention mechanism to improve the adaptiveness from the omni-scale, omni-dimension, and omni-process perspectives. Specifically, we leverage the meta-knowledge across the entire process, where Meta Triple Attention and Meta Temporal Pooling are presented respectively to adaptively capture omni-scale dependency from spatial/channel/temporal dimensions simultaneously and to adaptively aggregate temporal information through integrating the merits of three complementary temporal aggregation methods. Extensive experiments demonstrate the state-of-the-art performance of the proposed MetaGait. On CASIA-B, we achieve rank-1 accuracy of 98.7%, 96.0%, and 89.3% under three conditions, respectively. On OU-MVLP, we achieve rank-1 accuracy of 92.4%.

Hao Chen, Yunlong Yu, Yonghan Dong et al.

Few-Shot Segmentation (FSS) is challenging for limited support images and large intra-class appearance discrepancies. Most existing approaches focus on extracting high-level representations of the same layers for support-query correlations, neglecting the shift issue between different layers and scales, due to the huge difference between support and query samples. In this paper, we propose a Multi-Content Interaction Network (MCINet) to remedy this issue by fully exploiting and interacting with the multi-scale contextual information contained in the support-query pairs to supplement the same-layer correlations. Specifically, MCINet improves FSS from the perspectives of boosting the query representations by incorporating the low-level structural information from another query branch into the high-level semantic features, enhancing the support-query correlations by exploiting both the same-layer and adjacent-layer features, and refining the predicted results by a multi-scale mask prediction strategy, with which the different scale contents have bidirectionally interacted. Experiments on two benchmarks demonstrate that our approach reaches SOTA performances and outperforms the best competitors with many desirable advantages, especially on the challenging COCO dataset.

Fangtai Wu, Hailong Guo, Shijie Huang et al.

Customized image editing aims to equip pre-trained diffusion models with specific visual effects using limited paired data, typically via Low-Rank Adaptation (LoRA). As the number of desired effects grows, storing and dynamically loading numerous these effect LoRAs significantly increases deployment overhead. Furthermore, current pipelines typically cascade these effect LoRAs with acceleration modules for fast generation, which triggers severe parameter interference and results in concept bleeding and style degradation. We propose CollectionLoRA, a multi-teacher on-policy distillation framework capable of distilling the concepts of up to 50 different effect LoRAs along with few-step generation capabilities into a single LoRA. This fundamentally resolves the feature interference issue and significantly reduces deployment costs. Specifically, the method introduces (i) a Probabilistic Dual-Stream Routing mechanism that enables the model to randomly switch between data sources during training, effectively enhancing its generalization in unseen scenarios; (ii) an Asymmetric Orthogonal Prompting strategy to achieve concept isolation within the prompt space; (iii) a Coarse-to-Fine Distillation Objective to mitigate the distribution gap between the teacher and student models. Extensive evaluations show that CollectionLoRA distills all customized effects and few-step generation into a single LoRA, reducing deployment overhead while achieving concept fidelity comparable to or better than independently trained teacher models.

Mushui Liu, Jun Dan, Ziqian Lu et al.

Due to the large-scale image size and object variations, current CNN-based and Transformer-based approaches for remote sensing image semantic segmentation are suboptimal for capturing the long-range dependency or limited to the complex computational complexity. In this paper, we propose CM-UNet, comprising a CNN-based encoder for extracting local image features and a Mamba-based decoder for aggregating and integrating global information, facilitating efficient semantic segmentation of remote sensing images. Specifically, a CSMamba block is introduced to build the core segmentation decoder, which employs channel and spatial attention as the gate activation condition of the vanilla Mamba to enhance the feature interaction and global-local information fusion. Moreover, to further refine the output features from the CNN encoder, a Multi-Scale Attention Aggregation (MSAA) module is employed to merge the different scale features. By integrating the CSMamba block and MSAA module, CM-UNet effectively captures the long-range dependencies and multi-scale global contextual information of large-scale remote-sensing images. Experimental results obtained on three benchmarks indicate that the proposed CM-UNet outperforms existing methods in various performance metrics. The codes are available at https://github.com/XiaoBuL/CM-UNet.

Mushui Liu, Fangtai Wu, Bozheng Li et al.

Few-shot learning (FSL) aims to recognize new concepts using a limited number of visual samples. Existing approaches attempt to incorporate semantic information into the limited visual data for category understanding. However, these methods often enrich class-level feature representations with abstract category names, failing to capture the nuanced features essential for effective generalization. To address this issue, we propose a novel framework for FSL, which incorporates both the abstract class semantics and the concrete class entities extracted from Large Language Models (LLMs), to enhance the representation of the class prototypes. Specifically, our framework composes a Semantic-guided Visual Pattern Extraction (SVPE) module and a Prototype-Calibration (PC) module, where the SVPE meticulously extracts semantic-aware visual patterns across diverse scales, while the PC module seamlessly integrates these patterns to refine the visual prototype, enhancing its representativeness. Extensive experiments on four few-shot classification benchmarks and the BSCD-FSL cross-domain benchmarks showcase remarkable advancements over the current state-of-the-art methods. Notably, for the challenging one-shot setting, our approach, utilizing the ResNet-12 backbone, achieves an impressive average improvement of 1.95% over the second-best competitor.

Bozheng Li, Mushui Liu, Gaoang Wang et al.

In this paper, we propose a novel Temporal Sequence-Aware Model (TSAM) for few-shot action recognition (FSAR), which incorporates a sequential perceiver adapter into the pre-training framework, to integrate both the spatial information and the sequential temporal dynamics into the feature embeddings. Different from the existing fine-tuning approaches that capture temporal information by exploring the relationships among all the frames, our perceiver-based adapter recurrently captures the sequential dynamics alongside the timeline, which could perceive the order change. To obtain the discriminative representations for each class, we extend a textual corpus for each class derived from the large language models (LLMs) and enrich the visual prototypes by integrating the contextual semantic information. Besides, We introduce an unbalanced optimal transport strategy for feature matching that mitigates the impact of class-unrelated features, thereby facilitating more effective decision-making. Experimental results on five FSAR datasets demonstrate that our method set a new benchmark, beating the second-best competitors with large margins.

Mushui Liu, Bozheng Li, Yunlong Yu

Prompt tuning, which involves training a small set of parameters, effectively enhances the pre-trained Vision-Language Models (VLMs) to downstream tasks. However, they often come at the cost of flexibility and adaptability when the tuned models are applied to different datasets or domains. In this paper, we explore capturing the task-specific information via meticulous refinement of entire VLMs, with minimal parameter adjustments. When fine-tuning the entire VLMs for specific tasks under limited supervision, overfitting and catastrophic forgetting become the defacto factors. To mitigate these issues, we propose a framework named CLIP-CITE via designing a discriminative visual-text task, further aligning the visual-text semantics in a supervision manner, and integrating knowledge distillation techniques to preserve the gained knowledge. Extensive experimental results under few-shot learning, base-to-new generalization, domain generalization, and cross-domain generalization settings, demonstrate that our method effectively enhances the performance on specific tasks under limited supervision while preserving the versatility of the VLMs on other datasets.

Jingren Liu, Zhong Ji, YunLong Yu et al.

Parameter-efficient fine-tuning for continual learning (PEFT-CL) has shown promise in adapting pre-trained models to sequential tasks while mitigating catastrophic forgetting problem. However, understanding the mechanisms that dictate continual performance in this paradigm remains elusive. To unravel this mystery, we undertake a rigorous analysis of PEFT-CL dynamics to derive relevant metrics for continual scenarios using Neural Tangent Kernel (NTK) theory. With the aid of NTK as a mathematical analysis tool, we recast the challenge of test-time forgetting into the quantifiable generalization gaps during training, identifying three key factors that influence these gaps and the performance of PEFT-CL: training sample size, task-level feature orthogonality, and regularization. To address these challenges, we introduce NTK-CL, a novel framework that eliminates task-specific parameter storage while adaptively generating task-relevant features. Aligning with theoretical guidance, NTK-CL triples the feature representation of each sample, theoretically and empirically reducing the magnitude of both task-interplay and task-specific generalization gaps. Grounded in NTK analysis, our framework imposes an adaptive exponential moving average mechanism and constraints on task-level feature orthogonality, maintaining intra-task NTK forms while attenuating inter-task NTK forms. Ultimately, by fine-tuning optimizable parameters with appropriate regularization, NTK-CL achieves state-of-the-art performance on established PEFT-CL benchmarks. This work provides a theoretical foundation for understanding and improving PEFT-CL models, offering insights into the interplay between feature representation, task orthogonality, and generalization, contributing to the development of more efficient continual learning systems.

Yike Yuan, Xinghe Fu, Yunlong Yu et al.

In this paper, we propose a simple yet effective transformer framework for self-supervised learning called DenseDINO to learn dense visual representations. To exploit the spatial information that the dense prediction tasks require but neglected by the existing self-supervised transformers, we introduce point-level supervision across views in a novel token-based way. Specifically, DenseDINO introduces some extra input tokens called reference tokens to match the point-level features with the position prior. With the reference token, the model could maintain spatial consistency and deal with multi-object complex scene images, thus generalizing better on dense prediction tasks. Compared with the vanilla DINO, our approach obtains competitive performance when evaluated on classification in ImageNet and achieves a large margin (+7.2% mIoU) improvement in semantic segmentation on PascalVOC under the linear probing protocol for segmentation.

Weijie He, Mushui Liu, Yunlong Yu

Single-frame infrared small target (SIRST) detection poses a significant challenge due to the requirement to discern minute targets amidst complex infrared background clutter. In this paper, we focus on a weakly-supervised paradigm to obtain high-quality pseudo masks from the point-level annotation by integrating a novel learning-free method with the hybrid of the learning-based method. The learning-free method adheres to a sequential process, progressing from a point annotation to the bounding box that encompasses the target, and subsequently to detailed pseudo masks, while the hybrid is achieved through filtering out false alarms and retrieving missed detections in the network's prediction to provide a reliable supplement for learning-free masks. The experimental results show that our learning-free method generates pseudo masks with an average Intersection over Union (IoU) that is 4.3% higher than the second-best learning-free competitor across three datasets, while the hybrid learning-based method further enhances the quality of pseudo masks, achieving an additional average IoU increase of 3.4%.

Ziqian Lu, Qinyue Tong, Jun Liu et al.

Despite recent advances in MLLM-based medical image segmentation, existing LISA-like methods cannot reliably reject false queries and often produce hallucinated segmentation masks for absent targets. This limitation reduces practical reliability in both medical education and clinical use. In this work, we propose MedVeriSeg, a training-free verification framework that equips LISA-like medical segmentation models with the ability to identify and reject false queries which contain non-existent targets. Our key observation is that the similarity map between the [SEG] token feature and MLLM image features exhibits markedly different distribution patterns for true and false queries. Based on this, we introduce a Similarity Response Quality Scoring Module that characterizes the similarity map from three aspects: strength, compactness, and purity, producing an initial target-existence prediction. We further incorporate qualitative visual evidence by using GPT-4o to jointly assess the similarity heatmap and the results of Similarity Response Quality Scoring Module for final verification. Experiments on a small-scale benchmark constructed from SA-Med2D-20M show that MedVeriSeg effectively rejects false-query segmentation requests while maintaining reliable recognition of true queries.

Jiacheng Ying, Mushui Liu, Zhe Wu et al.

Blind face restoration has made great progress in producing high-quality and lifelike images. Yet it remains challenging to preserve the ID information especially when the degradation is heavy. Current reference-guided face restoration approaches either require face alignment or personalized test-tuning, which are unfaithful or time-consuming. In this paper, we propose a tuning-free method named RestorerID that incorporates ID preservation during face restoration. RestorerID is a diffusion model-based method that restores low-quality images with varying levels of degradation by using a single reference image. To achieve this, we propose a unified framework to combine the ID injection with the base blind face restoration model. In addition, we design a novel Face ID Rebalancing Adapter (FIR-Adapter) to tackle the problems of content unconsistency and contours misalignment that are caused by information conflicts between the low-quality input and reference image. Furthermore, by employing an Adaptive ID-Scale Adjusting strategy, RestorerID can produce superior restored images across various levels of degradation. Experimental results on the Celeb-Ref dataset and real-world scenarios demonstrate that RestorerID effectively delivers high-quality face restoration with ID preservation, achieving a superior performance compared to the test-tuning approaches and other reference-guided ones. The code of RestorerID is available at \url{https://github.com/YingJiacheng/RestorerID}.

Weijie He, Mushui Liu, Yunlong Yu et al.

Compositional text-to-video generation, which requires synthesizing dynamic scenes with multiple interacting entities and precise spatial-temporal relationships, remains a critical challenge for diffusion-based models. Existing methods struggle with layout discontinuity, entity identity drift, and implausible interaction dynamics due to unconstrained cross-attention mechanisms and inadequate physics-aware reasoning. To address these limitations, we propose DyST-XL, a \textbf{training-free} framework that enhances off-the-shelf text-to-video models (e.g., CogVideoX-5B) through frame-aware control. DyST-XL integrates three key innovations: (1) A Dynamic Layout Planner that leverages large language models (LLMs) to parse input prompts into entity-attribute graphs and generates physics-aware keyframe layouts, with intermediate frames interpolated via trajectory optimization; (2) A Dual-Prompt Controlled Attention Mechanism that enforces localized text-video alignment through frame-aware attention masking, achieving precise control over individual entities; and (3) An Entity-Consistency Constraint strategy that propagates first-frame feature embeddings to subsequent frames during denoising, preserving object identity without manual annotation. Experiments demonstrate that DyST-XL excels in compositional text-to-video generation, significantly improving performance on complex prompts and bridging a crucial gap in training-free video synthesis. The code is released in https://github.com/XiaoBuL/DyST-XL.

Wensheng Wu, Zheming Lu, Ziqian Lu et al.

Industrial anomaly detection faces significant challenges due to the scarcity of anomalous samples and the complexity of real-world anomalies. In this paper, we propose a foundation model-based anomaly synthesis pipeline (FMAS) that generates highly realistic anomalous samples without fine-tuning or class-specific training. Motivated by the distinct frequency-domain characteristics of anomalies, we introduce aWavelet Domain Attention Module (WDAM), which exploits adaptive sub-band processing to enhance anomaly feature extraction. The combination of FMAS and WDAM significantly improves anomaly detection sensitivity while maintaining computational efficiency. Comprehensive experiments on MVTec AD and VisA datasets demonstrate that WDAM, as a plug-and-play module, achieves substantial performance gains against existing baselines.

Fangtai Wu, Mushui Liu, Weijie He et al.

The unified autoregressive (AR) model excels at multimodal understanding and generation, but its potential for customized image generation remains underexplored. Existing customized generation methods rely on full fine-tuning or adapters, making them costly and prone to overfitting or catastrophic forgetting. In this paper, we propose \textbf{CoAR}, a novel framework for injecting subject concepts into the unified AR models while keeping all pre-trained parameters completely frozen. CoAR learns effective, specific subject representations with only a minimal number of parameters using a Layerwise Multimodal Context Learning strategy. To address overfitting and language drift, we further introduce regularization that preserves the pre-trained distribution and anchors context tokens to improve subject fidelity and re-contextualization. Additionally, CoAR supports training-free subject customization in a user-provided style. Experiments demonstrate that CoAR achieves superior performance on both subject-driven personalization and style personalization, while delivering significant gains in computational and memory efficiency. Notably, CoAR tunes less than \textbf{0.05\%} of the parameters while achieving competitive performance compared to recent Proxy-Tuning. Code: https://github.com/KZF-kzf/CoAR

D. She, Mushui Liu, Jingxuan Pang et al.

Customized generation has achieved significant progress in image synthesis, yet personalized video generation remains challenging due to temporal inconsistencies and quality degradation. In this paper, we introduce CustomVideoX, an innovative framework leveraging the video diffusion transformer for personalized video generation from a reference image. CustomVideoX capitalizes on pre-trained video networks by exclusively training the LoRA parameters to extract reference features, ensuring both efficiency and adaptability. To facilitate seamless interaction between the reference image and video content, we propose 3D Reference Attention, which enables direct and simultaneous engagement of reference image features with all video frames across spatial and temporal dimensions. To mitigate the excessive influence of reference image features and textual guidance on generated video content during inference, we implement the Time-Aware Reference Attention Bias (TAB) strategy, dynamically modulating reference bias over different time steps. Additionally, we introduce the Entity Region-Aware Enhancement (ERAE) module, aligning highly activated regions of key entity tokens with reference feature injection by adjusting attention bias. To thoroughly evaluate personalized video generation, we establish a new benchmark, VideoBench, comprising over 50 objects and 100 prompts for extensive assessment. Experimental results show that CustomVideoX significantly outperforms existing methods in terms of video consistency and quality.

Mushui Liu, Weijie He, Ziqian Lu et al.

Prompt learning is a powerful technique for transferring Vision-Language Models (VLMs) such as CLIP to downstream tasks. However, the prompt-based methods that are fine-tuned solely with base classes may struggle to generalize to novel classes in open-vocabulary scenarios, especially when data are limited. To address this issue, we propose an innovative approach called SYNC-CLIP that leverages SYNthetiC data for enhancing the generalization capability of CLIP. Based on the observation of the distribution shift between the real and synthetic samples, we treat real and synthetic samples as distinct domains and propose to optimize separate domain prompts to capture domain-specific information, along with the shared visual prompts to preserve the semantic consistency between two domains. By aligning the cross-domain features, the synthetic data from novel classes can provide implicit guidance to rebalance the decision boundaries. Experimental results on three model generalization tasks demonstrate that our method performs very competitively across various benchmarks. Notably, SYNC-CLIP outperforms the state-of-the-art competitor PromptSRC by an average improvement of 3.0% on novel classes across 11 datasets in open-vocabulary scenarios.

Hao Chen, Han Tao, Guo Song et al.

This paper presents Variables Adaptive Mixture of Experts (VAMoE), a novel framework for incremental weather forecasting that dynamically adapts to evolving spatiotemporal patterns in real time data. Traditional weather prediction models often struggle with exorbitant computational expenditure and the need to continuously update forecasts as new observations arrive. VAMoE addresses these challenges by leveraging a hybrid architecture of experts, where each expert specializes in capturing distinct subpatterns of atmospheric variables (temperature, humidity, wind speed). Moreover, the proposed method employs a variable adaptive gating mechanism to dynamically select and combine relevant experts based on the input context, enabling efficient knowledge distillation and parameter sharing. This design significantly reduces computational overhead while maintaining high forecast accuracy. Experiments on real world ERA5 dataset demonstrate that VAMoE performs comparable against SoTA models in both short term (1 days) and long term (5 days) forecasting tasks, with only about 25% of trainable parameters and 50% of the initial training data.

Jinzhe Jiang, Yaqian Zhao, Xin Zhang et al.

This paper explores the application of the parameter-shift rule (PSR) for computing gradients in unitary optical neural networks (UONNs). While backpropagation has been fundamental to training conventional neural networks, its implementation in optical neural networks faces significant challenges due to the physical constraints of optical systems. We demonstrate how PSR, which calculates gradients by evaluating functions at shifted parameter values, can be effectively adapted for training UONNs constructed from Mach-Zehnder interferometer meshes. The method leverages the inherent Fourier series nature of optical interference in these systems to compute exact analytical gradients directly from hardware measurements. This approach offers a promising alternative to traditional in silico training methods and circumvents the limitations of both finite difference approximations and all-optical backpropagation implementations. We present the theoretical framework and practical methodology for applying PSR to optimize phase parameters in optical neural networks, potentially advancing the development of efficient hardware-based training strategies for optical computing systems.

Mushui Liu, Yuhang Ma, Yang Zhen et al.

Diffusion models have exhibited substantial success in text-to-image generation. However, they often encounter challenges when dealing with complex and dense prompts involving multiple objects, attribute binding, and long descriptions. In this paper, we propose a novel framework called \textbf{LLM4GEN}, which enhances the semantic understanding of text-to-image diffusion models by leveraging the representation of Large Language Models (LLMs). It can be seamlessly incorporated into various diffusion models as a plug-and-play component. A specially designed Cross-Adapter Module (CAM) integrates the original text features of text-to-image models with LLM features, thereby enhancing text-to-image generation. Additionally, to facilitate and correct entity-attribute relationships in text prompts, we develop an entity-guided regularization loss to further improve generation performance. We also introduce DensePrompts, which contains $7,000$ dense prompts to provide a comprehensive evaluation for the text-to-image generation task. Experiments indicate that LLM4GEN significantly improves the semantic alignment of SD1.5 and SDXL, demonstrating increases of 9.69\% and 12.90\% in color on T2I-CompBench, respectively. Moreover, it surpasses existing models in terms of sample quality, image-text alignment, and human evaluation.

Jingren Liu, Zhong Ji, Yanwei Pang et al.

The proliferation of Few-Shot Class Incremental Learning (FSCIL) methodologies has highlighted the critical challenge of maintaining robust anti-amnesia capabilities in FSCIL learners. In this paper, we present a novel conceptualization of anti-amnesia in terms of mathematical generalization, leveraging the Neural Tangent Kernel (NTK) perspective. Our method focuses on two key aspects: ensuring optimal NTK convergence and minimizing NTK-related generalization loss, which serve as the theoretical foundation for cross-task generalization. To achieve global NTK convergence, we introduce a principled meta-learning mechanism that guides optimization within an expanded network architecture. Concurrently, to reduce the NTK-related generalization loss, we systematically optimize its constituent factors. Specifically, we initiate self-supervised pre-training on the base session to enhance NTK-related generalization potential. These self-supervised weights are then carefully refined through curricular alignment, followed by the application of dual NTK regularization tailored specifically for both convolutional and linear layers. Through the combined effects of these measures, our network acquires robust NTK properties, ensuring optimal convergence and stability of the NTK matrix and minimizing the NTK-related generalization loss, significantly enhancing its theoretical generalization. On popular FSCIL benchmark datasets, our NTK-FSCIL surpasses contemporary state-of-the-art approaches, elevating end-session accuracy by 2.9\% to 9.3\%.

Yunlong Yu, Zhong Ji, Zhongfei Zhang et al.

We introduce a simple yet effective episode-based training framework for zero-shot learning (ZSL), where the learning system requires to recognize unseen classes given only the corresponding class semantics. During training, the model is trained within a collection of episodes, each of which is designed to simulate a zero-shot classification task. Through training multiple episodes, the model progressively accumulates ensemble experiences on predicting the mimetic unseen classes, which will generalize well on the real unseen classes. Based on this training framework, we propose a novel generative model that synthesizes visual prototypes conditioned on the class semantic prototypes. The proposed model aligns the visual-semantic interactions by formulating both the visual prototype generation and the class semantic inference into an adversarial framework paired with a parameter-economic Multi-modal Cross-Entropy Loss to capture the discriminative information. Extensive experiments on four datasets under both traditional ZSL and generalized ZSL tasks show that our model outperforms the state-of-the-art approaches by large margins.

Zhong Ji, Xuejie Yu, Yunlong Yu et al.

Zero-Shot Classification (ZSC) equips the learned model with the ability to recognize the visual instances from the novel classes via constructing the interactions between the visual and the semantic modalities. In contrast to the traditional image classification, ZSC is easily suffered from the class-imbalance issue since it is more concerned with the class-level knowledge transfer capability. In the real world, the class samples follow a long-tailed distribution, and the discriminative information in the sample-scarce seen classes is hard to be transferred to the related unseen classes in the traditional batch-based training manner, which degrades the overall generalization ability a lot. Towards alleviating the class imbalance issue in ZSC, we propose a sample-balanced training process to encourage all training classes to contribute equally to the learned model. Specifically, we randomly select the same number of images from each class across all training classes to form a training batch to ensure that the sample-scarce classes contribute equally as those classes with sufficient samples during each iteration. Considering that the instances from the same class differ in class representativeness, we further develop an efficient semantics-guided feature fusion model to obtain discriminative class visual prototype for the following visual-semantic interaction process via distributing different weights to the selected samples based on their class representativeness. Extensive experiments on three imbalanced ZSC benchmark datasets for both the Traditional ZSC (TZSC) and the Generalized ZSC (GZSC) tasks demonstrate our approach achieves promising results especially for the unseen categories those are closely related to the sample-scarce seen categories.

Yunlong Yu, Zhong Ji, Yanwei Pang et al.

Existing generative Zero-Shot Learning (ZSL) methods only consider the unidirectional alignment from the class semantics to the visual features while ignoring the alignment from the visual features to the class semantics, which fails to construct the visual-semantic interactions well. In this paper, we propose to synthesize visual features based on an auto-encoder framework paired with bi-adversarial networks respectively for visual and semantic modalities to reinforce the visual-semantic interactions with a bi-directional alignment, which ensures the synthesized visual features to fit the real visual distribution and to be highly related to the semantics. The encoder aims at synthesizing real-like visual features while the decoder forces both the real and the synthesized visual features to be more related to the class semantics. To further capture the discriminative information of the synthesized visual features, both the real and synthesized visual features are forced to be classified into the correct classes via a classification network. Experimental results on four benchmark datasets show that the proposed approach is particularly competitive on both the traditional ZSL and the generalized ZSL tasks.

Yunlong Yu, Zhong Ji, Yanwei Fu et al.

Zero-Shot Learning (ZSL) is achieved via aligning the semantic relationships between the global image feature vector and the corresponding class semantic descriptions. However, using the global features to represent fine-grained images may lead to sub-optimal results since they neglect the discriminative differences of local regions. Besides, different regions contain distinct discriminative information. The important regions should contribute more to the prediction. To this end, we propose a novel stacked semantics-guided attention (S2GA) model to obtain semantic relevant features by using individual class semantic features to progressively guide the visual features to generate an attention map for weighting the importance of different local regions. Feeding both the integrated visual features and the class semantic features into a multi-class classification architecture, the proposed framework can be trained end-to-end. Extensive experimental results on CUB and NABird datasets show that the proposed approach has a consistent improvement on both fine-grained zero-shot classification and retrieval tasks.

Zhong Ji, Yuxin Sun, Yunlong Yu et al.

Zero-Shot Hashing aims at learning a hashing model that is trained only by instances from seen categories but can generate well to those of unseen categories. Typically, it is achieved by utilizing a semantic embedding space to transfer knowledge from seen domain to unseen domain. Existing efforts mainly focus on single-modal retrieval task, especially Image-Based Image Retrieval (IBIR). However, as a highlighted research topic in the field of hashing, cross-modal retrieval is more common in real world applications. To address the Cross-Modal Zero-Shot Hashing (CMZSH) retrieval task, we propose a novel Attribute-Guided Network (AgNet), which can perform not only IBIR, but also Text-Based Image Retrieval (TBIR). In particular, AgNet aligns different modal data into a semantically rich attribute space, which bridges the gap caused by modality heterogeneity and zero-shot setting. We also design an effective strategy that exploits the attribute to guide the generation of hash codes for image and text within the same network. Extensive experimental results on three benchmark datasets (AwA, SUN, and ImageNet) demonstrate the superiority of AgNet on both cross-modal and single-modal zero-shot image retrieval tasks.

Yunlong Yu, Zhong Ji, Jichang Guo et al.

Zero-Shot Learning (ZSL) is typically achieved by resorting to a class semantic embedding space to transfer the knowledge from the seen classes to unseen ones. Capturing the common semantic characteristics between the visual modality and the class semantic modality (e.g., attributes or word vector) is a key to the success of ZSL. In this paper, we propose a novel encoder-decoder approach, namely Latent Space Encoding (LSE), to connect the semantic relations of different modalities. Instead of requiring a projection function to transfer information across different modalities like most previous work, LSE per- forms the interactions of different modalities via a feature aware latent space, which is learned in an implicit way. Specifically, different modalities are modeled separately but optimized jointly. For each modality, an encoder-decoder framework is performed to learn a feature aware latent space via jointly maximizing the recoverability of the original space from the latent space and the predictability of the latent space from the original space. To relate different modalities together, their features referring to the same concept are enforced to share the same latent codings. In this way, the common semantic characteristics of different modalities are generalized with the latent representations. Another property of the proposed approach is that it is easily extended to more modalities. Extensive experimental results on four benchmark datasets (AwA, CUB, aPY, and ImageNet) clearly demonstrate the superiority of the proposed approach on several ZSL tasks, including traditional ZSL, generalized ZSL, and zero-shot retrieval (ZSR).

Yunlong Yu, Zhong Ji, Jichang Guo et al.

Zero-shot learning (ZSL) endows the computer vision system with the inferential capability to recognize instances of a new category that has never seen before. Two fundamental challenges in it are visual-semantic embedding and domain adaptation in cross-modality learning and unseen class prediction steps, respectively. To address both challenges, this paper presents two corresponding methods named Adaptive STructural Embedding (ASTE) and Self-PAsed Selective Strategy (SPASS), respectively. Specifically, ASTE formulates the visualsemantic interactions in a latent structural SVM framework to adaptively adjust the slack variables to embody the different reliableness among training instances. In this way, the reliable instances are imposed with small punishments, wheras the less reliable instances are imposed with more severe punishments. Thus, it ensures a more discriminative embedding. On the other hand, SPASS offers a framework to alleviate the domain shift problem in ZSL, which exploits the unseen data in an easy to hard fashion. Particularly, SPASS borrows the idea from selfpaced learning by iteratively selecting the unseen instances from reliable to less reliable to gradually adapt the knowledge from the seen domain to the unseen domain. Subsequently, by combining SPASS and ASTE, we present a self-paced Transductive ASTE (TASTE) method to progressively reinforce the classification capacity. Extensive experiments on three benchmark datasets (i.e., AwA, CUB, and aPY) demonstrate the superiorities of ASTE and TASTE. Furthermore, we also propose a fast training (FT) strategy to improve the efficiency of most of existing ZSL methods. The FT strategy is surprisingly simple and general enough, which can speed up the training time of most existing methods by 4~300 times while holding the previous performance.

Yunlong Yu, Zhong Ji, Xi Li et al.

As an important and challenging problem in computer vision, zero-shot learning (ZSL) aims at automatically recognizing the instances from unseen object classes without training data. To address this problem, ZSL is usually carried out in the following two aspects: 1) capturing the domain distribution connections between seen classes data and unseen classes data; and 2) modeling the semantic interactions between the image feature space and the label embedding space. Motivated by these observations, we propose a bidirectional mapping based semantic relationship modeling scheme that seeks for crossmodal knowledge transfer by simultaneously projecting the image features and label embeddings into a common latent space. Namely, we have a bidirectional connection relationship that takes place from the image feature space to the latent space as well as from the label embedding space to the latent space. To deal with the domain shift problem, we further present a transductive learning approach that formulates the class prediction problem in an iterative refining process, where the object classification capacity is progressively reinforced through bootstrapping-based model updating over highly reliable instances. Experimental results on three benchmark datasets (AwA, CUB and SUN) demonstrate the effectiveness of the proposed approach against the state-of-the-art approaches.