Vision Transformer

Sparse Autoencoders (SAEs) have demonstrated significant success in interpreting Large Language Models (LLMs) by decomposing dense representations into sparse, semantic components. However, their potential for analyzing Vision Transformers (ViTs) remains largely under-explored. In this work, we present the first application of SAEs to the ViT [CLS] token for out-of-distribution (OOD) detection, addressing the limitation of existing methods that rely on entangled feature representations. We propose a novel framework utilizing a Top-k SAE to disentangle the dense [CLS] features into a structured latent space. Through this analysis, we reveal that in-distribution (ID) data exhibits consistent, class-specific activation patterns, which we formalize as Class Activation Profiles (CAPs). Our study uncovers a key structural invariant: while ID samples preserve a stable pattern within CAPs, OOD samples systematically disrupt this structure. Leveraging this insight, we introduce a scoring function based on the divergence of core energy profiles to quantify the deviation from ideal activation profiles. Our method achieves strong results on the FPR95 metric, critical for safety-sensitive applications across multiple benchmarks, while also achieving competitive AUROC. Overall, our findings demonstrate that the sparse, disentangled features revealed by SAEs can serve as a powerful, interpretable tool for robust OOD detection in vision models.

A privacy-preserving clothing classification scheme is presented to enable secure occupant-centric control (OCC) systems. Although the utilization of camera images for HVAC control has been widely studied to optimize thermal comfort, privacy protection of occupant images has not been considered in prior works. While various privacy-preserving methods have been proposed for image classification, applying conventional schemes results in severe accuracy degradation. In this paper, we introduce a privacy-preserving classification method using Vision Transformer (ViT) applied to clothing insulation estimation. In an experiment using the DeepFashion dataset categorized by clothing insulation, while the conventional pixel-based method suffers a severe accuracy drop, our scheme maintains a high accuracy on encrypted images, showing no degradation from plain images across all categories.

FlashAttention improves efficiency through tiling, but its online softmax still relies on floating-point arithmetic for numerical stability, making full quantization difficult. We identify three main obstacles to integer-only FlashAttention: (1) scale explosion during tile-wise accumulation, (2) inefficient shift-based exponential operations on GPUs, and (3) quantization granularity constraints requiring uniform scales for integer comparison. To address these challenges, we propose \textit{QFlash}, an end-to-end integer FlashAttention design that performs softmax entirely in the integer domain and runs as a single Triton kernel. On seven attention workloads from ViT, DeiT, and Swin models, QFlash achieves up to 6.73$\times$ speedup over I-ViT and up to 8.69$\times$ speedup on Swin, while reducing energy consumption by 18.8% compared to FP16 FlashAttention, without sacrificing Top-1 accuracy on ViT/DeiT and remaining competitive on Swin under per-tensor quantization. Our code is publicly available at https://github.com/EfficientCompLab/qflash.

Vision Transformers trained only on image classification routinely transfer to tasks that demand spatial understanding, yet they receive no spatial supervision during pretraining. We ask where and how robustly such structure is encoded. Probing a frozen ViT-B/16 layerwise for two complementary properties, local patch boundaries (BSDS500) and per-patch depth (NYU Depth V2), reveals a clear hierarchy: boundary structure becomes linearly decodable at layers 5-6 (AP = 0.833), while depth, which requires integrating global cues, peaks two to three layers later at layer 8 (MAE = 0.0875). Both signals collapse at the final classification layer, and random-weight controls confirm the encodings are learned rather than architectural. Causal interventions add specificity: ablating the single direction a linear depth probe reads degrades depth decoding by up to 165%, while ablating any other direction changes it by less than 1%. Targeted activation patching along that direction shows the depth signal is partially re-derived at each layer rather than passively carried in the residual stream, with mid-layer interventions persisting most strongly downstream. The result is that a classification-trained ViT develops an actively maintained spatial hierarchy that mirrors the early-to-late progression observed in the primate visual cortex.

Spiking Vision Transformers (S-ViTs) offer a promising framework for energy-efficient visual learning. However, existing designs remain limited by two fundamental issues: the restricted information capacity of binary spike coding and the dense token interactions introduced by global self-attention. To address these challenges, this work proposes BSViT, a burst spiking-driven Vision Transformer featuring a Dual-Channel Burst Spiking Self-Attention (DBSSA) mechanism. DBSSA encodes queries with binary spikes and keys with burst spikes to enhance representational capacity. The value pathway adopts dual excitatory and inhibitory binary channels, enabling signed modulation and richer spike interactions. Importantly, the entire attention operation preserves addition-only computation, ensuring compatibility with energy-efficient neuromorphic hardware. To further reduce spike activity and incorporate spatial priors, a patch adjacency masking strategy is introduced to restrict attention to local neighborhoods, resulting in structure-aware sparsity and reduced computational overhead. In addition, burst spike coding is systematically integrated across the network to increase spike-level representational capacity beyond conventional binary spiking. Extensive experiments on both static and event-based vision benchmarks demonstrate that BSViT consistently outperforms existing spiking Transformers in accuracy while maintaining competitive energy efficiency.

Early detection and classifying brain tumors using Magnetic Resonance Imaging (MRI) images is highly important but difficult to extract in medical images. Convolutional Neural Networks (CNNs) are good at capturing both local texture and spatial information whereas Vision Transformers (ViTs) are good at capturing long-range global dependencies. We propose a new hybrid architecture that combines a SqueezeNet-style CNN branch with a MobileViT-style global transformer branch, through an Adaptive Attention Gate mechanism, in this paper. The gate learns dynamically per-sample, per-feature weights to weight the contribution of each branch, allowing context-sensitive merging of local and global representations. The proposed model has a test accuracy of 97.60, a precision of 97.30, a recall of 97.50, an F1-score of 97.40, and a macro-average area under the curve (AUC) of 0.9946 with a trained and evaluated on the Brain Tumor MRI Dataset (Kaggle). These scores are higher than single CNN and ViT baselines, and current competitive fusion methods, showing that dynamic feature weighting is an effective way to classify medical images.

Self-supervised learning has achieved remarkable success in learning visual representations from clean data, yet remains challenging when clean observations are sparse or not available at all. In this paper, we demonstrate that pretrained vision models can be leveraged to learn distortion-robust representations, which can then be effectively applied to downstream tasks operating on distorted observations. In particular, we propose an asymmetric knowledge distillation framework in which both teacher and student are initialized from the same pretrained Vision Transformer but receive different views of each image: the teacher processes clean images, while the student sees their distorted versions. We introduce multi-level distillation that aligns global embeddings, patch-level features, and attention maps and show that the student is able to approximate clean-image representations despite never directly accessing clean data. We evaluate our approach on image classification tasks across several datasets and under various distortions, consistently outperforming existing alternatives for the same amount of human supervision.

Brain tumors represent one of the most critical neurological conditions, where early and accurate diagnosis is directly correlated with patient survival rates. Manual interpretation of Magnetic Resonance Imaging (MRI) scans is time-intensive, subject to inter-observer variability, and demands significant specialist expertise. This paper proposes a deep learning framework for automated four-class brain tumor classification distinguishing glioma, meningioma, pituitary tumor, and healthy brain tissue from a dataset of 7,023 MRI scans. The proposed system employs a Vision Transformer (ViT-B/16) pretrained on ImageNet-21k as the backbone, augmented with a clinically motivated preprocessing and training pipeline. Contrast Limited Adaptive Histogram Equalization (CLAHE) is applied to enhance local contrast and accentuate tumor boundaries invisible to standard normalization. A two-stage fine-tuning strategy is adopted: the classification head is warmed up with the backbone frozen, followed by full fine-tuning with discriminative learning rates. MixUp and CutMix augmentation is applied per batch to improve generalization. Exponential Moving Average (EMA) of weights and Test-Time Augmentation (TTA) further stabilize and boost performance. Attention Rollout visualization provides clinically interpretable heatmaps of the brain regions driving each prediction. The proposed model achieves a test accuracy of 99.29%, macro F1-score of 99.25%, and perfect recall on both healthy and meningioma classes, outperforming all CNN-based baselines

With the development of deep learning, ViT-based stereo matching methods have made significant progress due to their remarkable robustness and zero-shot ability. However, due to the limitations of ViTs in handling resolution sensitivity and their relative neglect of local information, the ability of ViT-based methods to predict details and handle arbitrary-resolution images is still weaker than that of CNN-based methods. To address these shortcomings, we propose MLG-Stereo, a systematic pipeline-level design that extends global modeling beyond the encoder stage. First, we propose a Multi-Granularity Feature Network to effectively balance global context and local geometric information, enabling comprehensive feature extraction from images of arbitrary resolution and bridging the gap between training and inference scales. Then, a Local-Global Cost Volume is constructed to capture both locally-correlated and global-aware matching information. Finally, a Local-Global Guided Recurrent Unit is introduced to iteratively optimize the disparity locally under the guidance of global information. Extensive experiments are conducted on multiple benchmark datasets, demonstrating that our MLG-Stereo exhibits highly competitive performance on the Middlebury and KITTI-2015 benchmarks compared to contemporaneous leading methods, and achieves outstanding results in the KITTI-2012 dataset.

Standard transformer architectures learn fixed slow-weight representations during training and lack mechanisms for rapid adaptation within an episode. In contrast, biological neural systems address this through fast synaptic updates that form transient associative memories during inference, a property known as Hebbian plasticity. In this paper, we conduct an empirical study of Hebbian Fast-Weight (HFW) modules integrated into multiple transformer backbones, including ViT-Small, DeiT-Small, and Swin-Tiny. We evaluate six model variants: ViT, DeiT, Swin, ViT-Hebbian, DeiT-Hebbian, and Swin-Hebbian on 5-way 1-shot and 5-way 5-shot classification tasks using the Omniglot benchmark under a Prototypical Network meta-learning framework. We propose a single module placement strategy for Swin-Tiny in which one HFW module is applied to the final stage feature map after all hierarchical stages have completed. This design avoids the training instability caused by placing separate Hebbian modules at each stage and achieves the highest test accuracy across all six models (96.2% at 1-shot; 99.2% at 5-shot), outperforming its non-Hebbian baseline by $+0.3$ percentage points at 1-shot. We analyze the interaction between Swin's shifted window inductive bias and episode-level Hebbian binding, discuss why per-block placement fails for ViT and DeiT variants in a low-data regime, and situate the results within the wider literature on fast and slow-weight meta-learning.

Vision Transformers (ViTs) have achieved remarkable success across vision tasks, yet recent studies show they remain vulnerable to backdoor attacks. Existing patch-wise attacks typically assume a single fixed trigger location during inference to maximize trigger attention. However, they overlook the self-attention mechanism in ViTs, which captures long-range dependencies across patches. In this work, we observe that a patch-wise trigger can achieve high attack effectiveness when activating backdoors across neighboring patches, a phenomenon we term the Trigger Radiating Effect (TRE). We further find that inter-patch trigger insertion during training can synergistically enhance TRE compared to single-patch insertion. Prior ViT-specific attacks that maximize trigger attention often sacrifice visual and attention stealthiness, making them detectable. Based on these insights, we propose PASTA, a twofold stealthy patch-wise backdoor attack in both pixel and attention domains. PASTA enables backdoor activation when the trigger is placed at arbitrary patches during inference. To achieve this, we introduce a multi-location trigger insertion strategy to enhance TRE. However, preserving stealthiness while maintaining strong TRE is challenging, as TRE is weakened under stealthy constraints. We therefore formulate a bi-level optimization problem and propose an adaptive backdoor learning framework, where the model and trigger iteratively adapt to each other to avoid local optima. Extensive experiments show that PASTA achieves 99.13% attack success rate across arbitrary patches on average, while significantly improving visual and attention stealthiness (144.43x and 18.68x) and robustness (2.79x) against state-of-the-art ViT defenses across four datasets, outperforming CNN- and ViT-based baselines.

For state-of-the-art image understanding, Vision Transformers (ViTs) have become the standard architecture but their processing diverges substantially from human attentional characteristics. We investigate whether this cognitive gap can be shrunk by fine-tuning the self-attention weights of Google's ViT-B/16 on human saliency fixation maps. To isolate the effects of semantically relevant signals from generic human supervision, the tuned model is compared against a shuffled control. Fine-tuning significantly improved alignment across five saliency metrics and induced three hallmark human-like biases: tuning reversed the baseline's anti-human large-object bias toward small-objects, amplified the animacy preference and diminished extreme attention entropy. Bayesian parity analysis provides decisive to very-strong evidence that this cognitive alignment comes at no cost to the model's original classification performance on in- (ImageNet), corrupted (ImageNet-C) and out-of-distribution (ObjectNet) benchmarks. An equivalent procedure applied to a ResNet-50 Convolutional Neural Network (CNN) instead degraded both alignment and accuracy, suggesting that the ViT's modular self-attention mechanism is uniquely suited for dissociating spatial priority from representational logic. These findings demonstrate that biologically grounded priors can be instilled as a free emergent property of human-aligned attention, to improve transformer interpretability.

Despite the remarkable success of Vision Transformers (ViTs) across a wide range of vision tasks, recent studies have revealed that they remain vulnerable to adversarial examples, much like Convolutional Neural Networks (CNNs). A common empirical defense strategy is adversarial training, yet the theoretical underpinnings of its robustness in ViTs remain largely unexplored. In this work, we present the first theoretical analysis of adversarial training under simplified ViT architectures. We show that, when trained under a signal-to-noise ratio that satisfies a certain condition and within a moderate perturbation budget, adversarial training enables ViTs to achieve nearly zero robust training loss and robust generalization error under certain regimes. Remarkably, this leads to strong generalization even in the presence of overfitting, a phenomenon known as \emph{benign overfitting}, previously only observed in CNNs (with adversarial training). Experiments on both synthetic and real-world datasets further validate our theoretical findings.

Face Image Quality Assessment is crucial for reliable face recognition systems, yet existing Vision Transformer-based approaches rely exclusively on final-layer representations, ignoring quality-relevant information captured at intermediate network depths. This paper presents the first comprehensive investigation of how intermediate representations within ViTs contribute to face quality assessment through early exit mechanisms and score fusion strategies. We systematically analyze all twelve transformer blocks of ViT-FIQA architectures, demonstrating that different depths capture distinct and complementary quality-relevant information, as evidenced by varying attention patterns and performance characteristics across network layers. We propose a score fusion framework that combines quality predictions from multiple transformer blocks without architectural modifications or additional training. Our early exit analysis reveals optimal performance-efficiency trade-offs, enabling significant computational savings while maintaining competitive performance. Through extensive evaluation across eight benchmark datasets using four FR models, we demonstrate that our fusion strategy improves upon single-exit approaches. Our proposed quality fusion approach employs depth-weighted averaging that assigns progressively higher importance to deeper transformer blocks, achieving the best quality assessment performance by effectively leveraging the hierarchical nature of feature learning in ViTs. Our work challenges the conventional wisdom that only deep features matter for face analysis, revealing that intermediate representations contain valuable information for quality assessment. The proposed framework offers practical benefits for real-world biometric systems by enabling adaptive computation based on resource constraints while maintaining competitive quality assessment capabilities.

Face Image Quality Assessment (FIQA) aims to assess the recognition utility of face samples and is essential for reliable face recognition (FR) systems. Existing approaches require computationally expensive procedures such as multiple forward passes, backpropagation, or additional training, and only recent work has focused on the use of Vision Transformers. Recent studies highlighted that these architectures inherently function as saliency learners with attention patterns naturally encoding spatial importance. This work proposes ATTN-FIQA, a novel training-free approach that investigates whether pre-softmax attention scores from pre-trained Vision Transformer-based face recognition models can serve as quality indicators. We hypothesize that attention magnitudes intrinsically encode quality: high-quality images with discriminative facial features enable strong query-key alignments producing focused, high-magnitude attention patterns, while degraded images generate diffuse, low-magnitude patterns. ATTN-FIQA extracts pre-softmax attention matrices from the final transformer block, aggregate multi-head attention information across all patches, and compute image-level quality scores through simple averaging, requiring only a single forward pass through pre-trained models without architectural modifications, backpropagation, or additional training. Through comprehensive evaluation across eight benchmark datasets and four FR models, this work demonstrates that attention-based quality scores effectively correlate with face image quality and provide spatial interpretability, revealing which facial regions contribute most to quality determination.

Vision Transformers $(\texttt{ViT})$ have become the architecture of choice for many computer vision tasks, yet their performance in computer-aided diagnostics remains limited. Focusing on breast cancer detection from mammograms, we identify two main causes for this shortfall. First, medical images are high-resolution with small abnormalities, leading to an excessive number of tokens and making it difficult for the softmax-based attention to localize and attend to relevant regions. Second, medical image classification is inherently fine-grained, with low inter-class and high intra-class variability, where standard cross-entropy training is insufficient. To overcome these challenges, we propose a framework with three key components: (1) Region of interest $(\texttt{RoI})$ based token reduction using an object detection model to guide attention; (2) contrastive learning between selected $\texttt{RoI}$ to enhance fine-grained discrimination through hard-negative based training; and (3) a $\texttt{DINOv2}$ pretrained $\texttt{ViT}$ that captures localization-aware, fine-grained features instead of global $\texttt{CLIP}$ representations. Experiments on public mammography datasets demonstrate that our method achieves superior performance over existing baselines, establishing its effectiveness and potential clinical utility for large-scale breast cancer screening. Our code is available for reproducibility here: https://aih-iitd.github.io/publications/attend-what-matters

Hyperspectral image (HSI) classification remains challenging due to high spectral dimensionality, redundancy, and limited labeled data. Although convolutional neural networks (CNNs) and Vision Transformers (ViTs) achieve strong performance by exploiting spectral-spatial information and long-range dependencies, they often incur high computational cost and large model size, limiting practical use. To address these limitations, a unified hybrid framework, termed ConvVitMamba, is proposed for efficient HSI classification. The architecture integrates three components: a multiscale convolutional feature extractor to capture local spectral, spatial, and joint patterns; a Vision Transformer based tokenization and encoding stage to model global contextual relationships; and a lightweight Mamba inspired gated sequence mixing module for efficient content-aware refinement without quadratic self-attention. Principal Component Analysis (PCA) is used as preprocessing to reduce redundancy and improve efficiency. Experiments on four benchmark datasets, including Houston and three UAV borne QUH datasets (Pingan, Qingyun, and Tangdaowan), demonstrate that ConvVitMamba consistently outperforms CNN, Transformer, and Mamba based methods while maintaining a favorable balance between accuracy, model size, and inference efficiency. Ablation studies confirm the complementary contributions of all components. The results indicate that the proposed framework provides an effective and efficient solution for HSI classification in diverse scenarios. The source code is publicly available at https://github.com/mqalkhatib/ConvVitMamba

Visual state-space models (SSMs) are increasingly promoted as efficient alternatives to Vision Transformers, yet their practical advantages remain unclear under fair comparison because existing studies rarely isolate encoder effects from decoder and training choices. We present a strictly controlled benchmark of representative visual SSM families, including VMamba, MambaVision, and Spatial-Mamba, for remote-sensing semantic segmentation, in which only the encoder varies across experiments. Evaluated on LoveDA and ISPRS Potsdam under a unified 4-stage feature interface and a fixed lightweight decoder, the benchmark reveals three main findings, intra-family scaling yields only modest gains, cross-domain generalization is strongly asymmetric, and boundary delineation is the dominant failure mode under distribution shift. Although visual SSMs achieve favorable accuracy-efficiency trade-offs relative to the controlled CNN and Transformer baselines considered here, the results suggest that future improvements are more likely to come from robustness-oriented design and boundary-aware decoding than from encoder scaling alone. By isolating encoder behavior under a unified and reproducible protocol, this study establishes a practical reference benchmark for the design and evaluation of future Mamba-based segmentation backbones

In recent years, the Vision Transformer (ViT) has garnered significant attention within the computer vision community. However, the core component of ViT, Self-Attention, lacks explicit spatial priors and suffers from quadratic computational complexity, limiting its applicability. To address these issues, we have proposed RMT, a robust vision backbone with explicit spatial priors for general purposes. RMT utilizes Manhattan distance decay to introduce spatial information and employs a horizontal and vertical decomposition attention method to model global information. Building on the strengths of RMT, Euclidean enhanced Vision Transformer (EVT) is an expanded version that incorporates several key improvements. Firstly, EVT uses a more reasonable Euclidean distance decay to enhance the modeling of spatial information, allowing for a more accurate representation of spatial relationships compared to the Manhattan distance used in RMT. Secondly, EVT abandons the decomposed attention mechanism featured in RMT and instead adopts a simpler spatially-independent grouping approach, providing the model with greater flexibility in controlling the number of tokens within each group. By addressing these modifications, EVT offers a more sophisticated and adaptable approach to incorporating spatial priors into the Self-Attention mechanism, thus overcoming some of the limitations associated with RMT and further enhancing its applicability in various computer vision tasks. Extensive experiments on Image Classification, Object Detection, Instance Segmentation, and Semantic Segmentation demonstrate that EVT exhibits exceptional performance. Without additional training data, EVT achieves 86.6% top1-acc on ImageNet-1k.

Vision Transformers (ViTs) have become a dominant architecture in computer vision, yet their prediction process remains difficult to interpret because information is propagated through complex interactions across layers and attention heads. Existing attention based explanation methods provide an intuitive way to trace information flow. However, they rely mainly on raw attention weights, which do not explicitly reflect the final decision and often lead to explanations with limited class discriminability. In contrast, gradient based localization methods are more effective at highlighting class specific evidence, but they do not fully exploit the hierarchical attention propagation mechanism of transformers. To address this limitation, we propose Decision-Aware Attention Propagation (DAP), an attribution method that injects decision-relevant priors into transformer attention propagation. By estimating token importance through gradient based localization and integrating it into layer wise attention rollout, the method captures both the structural flow of attention and the evidence most relevant to the final prediction. Consequently, DAP produces attribution maps that are more class sensitive, compact, and faithful than those generated by conventional attention based methods. Extensive experiments across Vision Transformer variants of different model scales show that DAP consistently outperforms existing baselines in both quantitative metrics and qualitative visualizations, indicating that decision aware propagation is an effective direction for improving ViT interpretability.

Plankton monitoring is essential for assessing aquatic ecosystems but is limited by the labor-intensive nature of manual microscopic analysis. Automating the segmentation of plankton from crowded images is crucial, however, it faces two major challenges: (i) the scarcity of pixel-level annotated datasets and (ii) the difficulty of distinguishing plankton from debris and overlapping individuals using conventional CNN-based methods. To address these issues, we propose PlankFormer, a novel framework for plankton instance segmentation. First, to overcome the data shortage, we introduce a method to generate labeled Pseudo Community Images (PCI) by synthesizing individual plankton images onto diverse backgrounds, including those created by generative models. Second, we propose a segmentation model utilizing a Vision Transformer (ViT) backbone with a Mask2Former decoder. To robustly capture the global structural features of plankton against occlusion and debris, we employ a Masked Autoencoder (MAE) for self-supervised pre-training on unlabeled individual images. Experimental results on real-world datasets demonstrate that our method significantly outperforms conventional methods, such as Mask R-CNN, particularly in challenging environments with high debris density. We demonstrate that our synthetic training strategy and MAE-based architecture enable high-precision segmentation with requiring less manual annotations for individual plankton images.

In today's day and age, we face a challenge in detecting deepfake images because of the fast evolution of modern generative models and the poor generalization capability of existing methods. In this paper, we use an ensemble of fine-tuned vision transformers like DINOv2, AIMv2 and OpenCLIP's ViT-L/14 to create generalizable method to detect deepfakes. We use the DF-Wild dataset released as part of the IEEE SP Cup 2025, because it uses a challenging and diverse set of manipulations and generation techniques. We started our experiments with CNN classifiers trained on spatial features. Experimental results show that our ensemble outperforms individual models and strong CNN baselines, achieving an AUC of 96.77% and an Equal Error Rate (EER) of just 9% on the DF-Wild test set, beating the state-of-the-art deepfake detection algorithm Effort by 7.05% and 8% in AUC and EER respectively. This was the winning solution for SP Cup, presented at ICASSP 2025.

Vision Transformer (ViT) has brought new breakthroughs to the field of image classification by introducing the self-attention mechanism and Graph Convolutional Networks(GCN) have been proposed and successfully applied in data representation and analysis. However, there are key challenges which limit their further development: (1) The patch size selected by ViT is crucial for accurate predictions, which raises a natural question: How to select the size of patches properly or how to comprehensively combine small patches and larger patches; (2) While the spatial structure information is important in vision tasks, the 1D position embeddings fails to capture the spatial structure information of patches more accurately; (3) The GCN can capture the local connectivity relationships between image nodes, but it lacks the ability to capture global graph structural information. On the contrary, the self-attention mechanism of ViT can draw the global relation on image patches, but it is unable to model the local structure of image. To overcome such limitations, we propose the Hierarchical Vision Transformer Enhanced by Graph Convolutional Network (GCN-HViT) for image classification. Specifically, the Hierarchical ViT we designed can model patch-wise information interactions on a global scale within each level and model hierarchical relationships between small patches and large patches across multiple levels. In addition, the proposed GCN method functions as a local feature extractor to obtain the local representation of each image patch which serves as a 2D position embedding of each patch in the 2D space. Meanwhile, it models patch-wise information interactions on a local scale within each level. Extensive experiments on 3 real-world datasets demonstrate that GCN-HViT achieves state-of-the-art performance.

This paper presents FairNVT, a lightweight debiasing framework for pretrained transformer-based encoders that improves both representation and prediction level fairness while preserving task accuracy. Unlike many existing debiasing approaches that address these notions separately, we argue they are inherently connected: suppressing sensitive information at the representation level can facilitate fairer predictions. Our approach learns task-relevant and sensitive embeddings via lightweight adapters, applies calibrated Gaussian noise to the sensitive embedding, and fuses it with the task representation. Together with orthogonality constraints and fairness regularization, these components jointly reduce sensitive-attribute leakage in the learned embeddings and encourage fairer downstream predictions. The framework is compatible with a wide range of pretrained transformer encoders. Across three datasets spanning vision and language, FairNVT reduces sensitive-attribute attacker accuracy, improves demographic-parity and equalized-odds metrics, and maintains high task performance.

We present a system for automated detection, localization, and scoring of arrow punctures on 40,cm indoor archery target faces, trained on only 48 annotated photographs (5{,}084 punctures). Our pipeline combines three components: a color-based canonical rectification stage that maps perspective-distorted photographs into a standardized coordinate system where pixel distances correspond to known physical measurements; a frozen self-supervised vision transformer (DINOv3 ViT-L/16) paired with AnyUp guided feature upsampling to recover sub-millimeter spatial precision from $32 \times 32$ patch tokens; and lightweight CenterNet-style detection heads for arrow-center heatmap prediction. Only 3.8,M of 308,M total parameters are trainable. Across three cross-validation folds, we achieve a mean F1 score of $0.893 \pm 0.011$ and a mean localization error of $1.41 \pm 0.06$,mm, comparable to or better than prior fully-supervised approaches that require substantially more training data. An ablation study shows that the CenterNet offset regression head, typically essential for sub-pixel refinement, provides negligible detection improvement while degrading localization in our setting. This suggests that guided feature upsampling already resolves the spatial precision lost through patch tokenization. On downstream archery metrics, the system recovers per-image average arrow scores with a median error of 1.8% and group centroid positions to within a median of 4.00,mm. These results demonstrate that frozen foundation models with minimal task-specific adaptation offer a practical paradigm for dense prediction in small-data regimes.

Time series forecasting plays a pivotal role in critical sectors such as finance, energy, transportation, and meteorology. However, Long-term Time Series Forecasting (LTSF) remains a significant challenge because real-world signals contain highly entangled temporal dynamics that are difficult to fully capture from a purely 1D perspective. To break this representation bottleneck, we propose TriTS, a novel cross-modal disentanglement framework that projects 1D time series into orthogonal time, frequency, and 2D-vision spaces.To seamlessly bridge the 1D-to-2D modality gap without the prohibitive $O(N^2)$ computational overhead of Vision Transformers (ViTs), we introduce a Period-Aware Reshaping strategy and incorporate Visual Mamba (Vim). This approach efficiently models cross-period dependencies as global visual textures while maintaining linear computational complexity. Complementing this, we design a Multi-Resolution Wavelet Mixing (MR-WM) module for the frequency modality, which explicitly decouples non-stationary signals into trend and noise components to achieve fine-grained time-frequency localization. Finally, a streaming linear branch is retained in the time domain to anchor numerical stability. By dynamically fusing these three complementary representations, TriTS effectively adapts to diverse data contexts. Extensive experiments across multiple benchmark datasets demonstrate that TriTS achieves state-of-the-art (SOTA) performance, fundamentally outperforming existing vision-based forecasters by drastically reducing both parameter count and inference latency.

Training-free token reduction methods for Vision Transformers (ToMe, ToFu, PiToMe, and MCTF) employ different scoring mechanisms, yet they share a closely matched cliff-like collapse at high compression. This paper explains \emph{why}. We develop a diagnostic framework with two tools, ranking consistency $ρ_s$ and off-diagonal correlation $ρ_\text{off}$, that decomposes the collapse into (1)a signal-agnostic error amplifier inherent to layer-wise reduction, predicting convex Pareto curves and $r_{\text{crit}} \propto 1/L$; and (2)shared reliance on \emph{pairwise} similarity signals whose ranking consistency degrades from $ρ_s{=}0.88$ to $0.27$ in deep layers. Pairwise rankings are inherently unstable ($O(N_p^2)$ joint perturbations) while unary signals enjoy greater stability ($O(N_p)$ perturbations, CLT). From three design principles derived from this diagnosis, we construct CATIS as a constructive validation: unary signals raise the trigger threshold, triage suppresses the gain. On ViT-Large at 63% FLOPs reduction, CATIS retains 96.9% of vanilla accuracy (81.0%) on ImageNet-1K where all baselines collapse to 43--65%.

Systolic murmurs are extra heart sounds that occur during the contraction phase of the cardiac cycle, often indicating heart abnormalities caused by turbulent blood flow. Their intensity, pitch, and quality vary, requiring precise identification for the accurate diagnosis of cardiac disorders. This study presents an automatic classification system for systolic murmurs using a feature extraction module, followed by a classification model. The feature extraction module employs complex orthogonal matching pursuit to project single or multiple murmur segments onto a redundant dictionary composed of multiresolution complex Gabor basis functions (GBFs). The resulting projection weights are split and reshaped into variable-resolution time--frequency feature matrices. Processing multiple segments of a single recording using a shared dictionary mitigates murmur variability. This is achieved by learning the weights for each segment while enforcing that they correspond to the same set of basis functions in the dictionary, promoting consistent time--frequency feature matrices. The classification model is built based on a vision transformer to process multiple input matrices of different resolutions by passing each through a convolutional neural network for patch tokenization. All embedding tokens are then concatenated to form a matrix and forwarded to an encoder layer that includes multihead attention, residual connections, and a convolutional network with a kernel size of one. This integration of multiresolution feature extraction with transformer-based feature classification enhances the accuracy and reliability of heart murmur identification. An experimental analysis of four types of systolic murmurs from the CirCor DigiScope dataset demonstrates the effectiveness of the system, achieving a classification accuracy of $95.96\%$.

Token pruning methods for Vision Transformers (ViTs) promise quadratic reductions in attention FLOPs by dropping uninformative patches. Yet standard variable-length attention APIs -- including FlashAttention-2's varlen and PyTorch's NestedTensor SDPA -- fail to translate these savings into proportional wall-clock gains at the short post-pruning sequence lengths typical of ViTs ($\leq$197 tokens). We identify a dispatch-overhead bottleneck: at these lengths, host-side kernel dispatch consumes ${\sim}$50,$μ$s regardless of workload, exceeding the actual GPU compute time at moderate-to-high pruning rates. We present a lightweight bidirectional Triton attention kernel whose dispatch floor is ${\sim}$24,$μ$s -- roughly 2.17$\times$ lower than FlashAttention-2 varlen -- allowing pruning savings to become visible in wall-clock time. Integrated into a complete pack-attend-unpack pipeline and evaluated on an NVIDIA RTX 4000 Ada Generation GPU, our system achieves 1.88$\times$ end-to-end throughput over padded PyTorch SDPA at standard 224$\times$224 inputs, scaling to 2.51$\times$ at 384$\times$384. Against FlashAttention-2 varlen -- the strongest baseline -- our kernel delivers 9-12% higher throughput at serving batch sizes (BS=1-4), and 2.17$\times$ lower kernel latency at 80% token pruning. Numerical correctness is verified with max absolute logit difference $<$0.004 and bit-exact top-1 predictions.

Existing single-modal RGB trackers often face performance bottlenecks in complex dynamic scenes, while the introduction of event sensors offers new potential for enhancing tracking capabilities. However, most current RGB-event fusion methods, primarily designed in the spatial domain using convolutional, Transformer, or Mamba architectures, fail to fully exploit the unique temporal response and high-frequency characteristics of event data. To address this, we1 propose FreqTrack, a frequency-aware RGBE tracking framework that establishes complementary inter-modal correlations through frequency-domain transformations for more robust feature fusion. We design a Spectral Enhancement Transformer (SET) layer that incorporates multi-head dynamic Fourier filtering to adaptively enhance and select frequency-domain features. Additionally, we develop a Wavelet Edge Refinement (WER) module, which leverages learnable wavelet transforms to explicitly extract multi-scale edge structures from event data, effectively improving modeling capability in high-speed and low-light scenarios. Extensive experiments on the COESOT and FE108 datasets demonstrate that FreqTrack achieves highly competitive performance, particularly attaining leading precision of 76.6% on the COESOT benchmark, validating the effectiveness of frequency-domain modeling for RGBE tracking.

Background: Task-specific microscopy datasets are often small, making it difficult to train deep learning models that learn robust features. While self-supervised learning (SSL) has shown promise through pretraining on large, domain-specific datasets, generalizability across datasets with differing staining protocols and channel configurations remains underexplored. We investigated the generalizability of SSL models pretrained on ImageNet-1k and HPA FOV, evaluating their embeddings on OpenCell with and without fine-tuning, two channel-mismatch strategies, and varying fine-tuning data fractions. We additionally analyzed single-cell embeddings on a labeled OpenCell subset. Result: DINO-based ViT backbones pretrained on HPA FOV or ImageNet-1k transfer well to OpenCell even without fine-tuning. Fine-tuning further improved performance to 0.704 $\pm$ 0.027 (17 classes, multi-class labels). At the single-cell level, the HPA single-cell-pretrained model achieved the highest k-nearest neighbor performance across all neighborhood sizes (macro $F_1$ $\geq$ 0.515). Conclusion: SSL methods like DINO, pretrained on large domain-relevant datasets, enable effective use of deep learning features for fine-tuning on small, task-specific microscopy datasets.

Spatiotemporal neural dynamics and oscillatory synchronization are widely implicated in biological information processing and have been hypothesized to support flexible coordination such as feature binding. By contrast, most deep learning architectures represent and propagate information through activation values, neglecting the joint dynamics of rate and phase. In this work, we introduce Kuramoto oscillatory Phase Encoding (KoPE) as an additional, evolving phase state to Vision Transformers, incorporating a neuro-inspired synchronization mechanism to advance learning efficiency. We show that KoPE can improve training, parameter, and data efficiency of vision models through synchronization-enhanced structure learning. Moreover, KoPE benefits tasks requiring structured understanding, including semantic and panoptic segmentation, representation alignment with language, and few-shot abstract visual reasoning (ARC-AGI). Theoretical analysis and empirical verification further suggest that KoPE can accelerate attention concentration for learning efficiency. These results indicate that synchronization can serve as a scalable, neuro-inspired mechanism for advancing state-of-the-art neural network models. Code is avaliable at https://github.com/microsoft/Neuro-inspired_Phase_Encoding.

Pretrained Vision Transformers (ViTs) such as DINOv2 and MAE provide generic image features that can be applied to a variety of downstream tasks such as retrieval, classification, and segmentation. However, such representations tend to focus on the most salient visual cues in the image, with no way to direct them toward less prominent concepts of interest. In contrast, Multimodal LLMs can be guided with textual prompts, but the resulting representations tend to be language-centric and lose their effectiveness for generic visual tasks. To address this, we introduce Steerable Visual Representations, a new class of visual representations, whose global and local features can be steered with natural language. While most vision-language models (e.g., CLIP) fuse text with visual features after encoding (late fusion), we inject text directly into the layers of the visual encoder (early fusion) via lightweight cross-attention. We introduce benchmarks for measuring representational steerability, and demonstrate that our steerable visual features can focus on any desired objects in an image while preserving the underlying representation quality. Our method also matches or outperforms dedicated approaches on anomaly detection and personalized object discrimination, exhibiting zero-shot generalization to out-of-distribution tasks.

Vision Transformers require significant computational resources and memory bandwidth, severely limiting their deployment on resource-constraint hardware. Most structured pruning methods reduce theoretical cost effectively, yet they typically operate at a single structural granularity and depend on multi-stage pipelines with importance ranking, auxiliary solvers or post-hoc magnitude thresholding, followed by a separate fine-tuning phase to recover accuracy. We propose Hierarchical Auto-Pruning (HiAP), which casts ViT pruning as a single budget-aware learning problem and jointly allocates sparsity across four granularities in one end-to-end phase. HiAP introduces stochastic Gumbel-Sigmoid gates at macro level (attention heads and FFN blocks) and micro level (intra-head dimensions and FFN neurons), and optimizes them against the task loss together with an analytical MAC cost term. The budget coefficient steers the network to a target compute level while the gates gradually harden into a dense, smaller sub-network at convergence. It does not require importance heuristics, ranking metrics, auxillary solvers or secondary fine-tuning. On ImageNet with DeiT small, HiAP automatically discovers hetergenous architectures, pruning depths, heads, and width by different amount across layers, and reaches competitive accuracies against substantially more complex pruning pipelines at comparable compute from a single training run.

Recent token merging techniques for Vision Transformers (ViTs) provide substantial speedups by reducing the number of tokens processed by self-attention, often without retraining. However, their direct application to the Segment Anything Model (SAM) family is nontrivial: SAM's image encoder mixes windowed and global attention, and its mask decoder relies on dense, prompt-conditioned features for precise boundary prediction. We systematically evaluate representative token-merging methods on SAM and Medical SAM in a strict off-the-shelf setting, and find that existing destination-selection heuristics can erode boundaries and leak prompt information as merge rates increase. We propose \textbf{StructSAM}, a resolution-preserving merge-unmerge framework tailored to SAM. StructSAM computes a lightweight token-energy score from first-order feature gradients, uses grid-based flatness screening to protect boundary and prompt regions, and merges tokens within flat areas toward low-energy destinations with explicit token recovery. We further provide a spectral graph coarsening view showing that score-guided merging yields bounded Laplacian spectral distortion compared to random or window-restricted baselines. Across eight natural and medical benchmarks, StructSAM reduces encoder FLOPs by 25-30% (up to 40%+ with prompt-aware merging) with minor drops in mIoU/Dice, consistently outperforming ToMe, PiToMe, ToMeSD, VidToMe, and ALGM at the same compute.

Conventional image denoising models often inadvertently learn spurious correlations between environmental factors and noise patterns. Moreover, due to high-frequency ambiguity, they struggle to reliably distinguish subtle textures from stochastic noise, resulting in over-removed details or residual noise artifacts. We therefore revisit denoising via causal intervention, arguing that purely correlational fitting entangles intrinsic content with extrinsic noise, which directly degrades robustness under distribution shifts. Motivated by this, we propose the Teacher-Guided Causal Disentanglement Network (TCD-Net), which explicitly decomposes the generative mechanism via structured interventions on feature spaces within a Vision Transformer framework. Specifically, our method integrates three key components: (1) An Environmental Bias Adjustment (EBA) module projects features into a stable, de-centered subspace to suppress global environmental bias (de-confounding). (2) A dual-branch disentanglement head employs an orthogonality constraint to force a strict separation between content and noise representations, preventing information leakage. (3) To resolve structural ambiguity, we leverage Nano Banana Pro, Google's reasoning-guided AI image generation model, to guide a causal prior, effectively pulling content representations back onto the natural-image manifold. Extensive experiments demonstrate that TCD-Net outperforms mainstream methods across multiple benchmarks in both fidelity and efficiency, achieving a real-time speed of 104.2 FPS on a single RTX 5090 GPU.

With the rise of large-scale foundation models, efficiently adapting them to downstream tasks remains a central challenge. Linear probing, which freezes the backbone and trains a lightweight head, is computationally efficient but often restricted to last-layer representations. We show that task-relevant information is distributed across the network hierarchy rather than encoded solely in the last layers. To leverage this distribution of information, we apply an attentive probing mechanism that dynamically fuses representations from all layers of a Vision Transformer. This attentive layer fusion (ALF) learns to identify the most relevant layers for a target task and combines low-level structural cues with high-level semantic abstractions. Across 20 diverse datasets and multiple pretrained foundation models, ALF achieves consistent, substantial gains over standard linear probes. Attention heatmaps further reveal that tasks different from the pre-training domain benefit most from intermediate representations. Overall, our findings underscore the value of intermediate layers and demonstrate a principled, task-aware approach for unlocking their potential for probing-based adaptation.

Contrastive Language-Image Pre-training (CLIP) relies on Vision Transformers whose attention mechanism is susceptible to spurious correlations, and scales quadratically with resolution. To address these limitations, We present CLIMP, the first fully Mamba-based contrastive vision-language model that replaces both the vision and text encoders with Mamba. The new architecture encodes sequential structure in both vision and language, with VMamba capturing visual spatial inductive biases, reducing reliance on spurious correlations and producing an embedding space favorable for cross-modal retrieval and out-of-distribution robustness-surpassing OpenAI's CLIP-ViT-B by 7.5% on ImageNet-O. CLIMP naturally supports variable input resolutions without positional encoding interpolation or specialized training, achieving up to 6.6% higher retrieval accuracy at 16x training resolution while using 5x less memory and 1.8x fewer FLOPs. The autoregressive text encoder further overcomes CLIP's fixed context limitation, enabling dense captioning retrieval. Our findings suggest that Mamba exhibits advantageous properties for vision-language learning, making it a compelling alternative to Transformer-based CLIP.The code and models are publicly available at https://github.com/NimrodShabtay/CLIMP}

Vision Transformers (ViTs) lack the hierarchical inductive biases inherent to Convolutional Neural Networks (CNNs), theoretically allowing them to maintain high-dimensional representations throughout all layers. However, recent observations suggest ViTs often spontaneously manifest a "U-shaped" entropy profile-compressing information in middle layers before expanding it for the final classification. In this work, we demonstrate that this "Inductive Bottleneck" is not an architectural artifact, but a data-dependent adaptation. By analyzing the layer-wise Effective Encoding Dimension (EED) of DINO-trained ViTs across datasets of varying compositional complexity (UC Merced, Tiny ImageNet, and CIFAR-100), we show that the depth of the bottleneck correlates strongly with the semantic abstraction required by the task. We find that while texture-heavy datasets preserve high-rank representations throughout, object-centric datasets drive the network to dampen high-frequency information in middle layers, effectively "learning" a bottleneck to isolate semantic features.

Input transformation-based attacks improve adversarial transferability by aggregating gradients over transformed inputs. Existing analyses mainly explain their efficacy from image diversity, semantic preservation, attention variance or hypothesis space augmentation, yet overlook the critical role of model frontend responses. In this paper, we revisit transformation-based attacks from an implicit ensemble perspective: each transformation can be viewed as a pre-processing operator before the surrogate model, inducing a distinct frontend response for gradient aggregation. Based on this view, we propose FRO, a Frontend Response-Oriented input transformation method that enriches such responses through two complementary operators. The Local Scaling Operator perturbs local content sampling via block-wise stretch-and-shrink operations, while the Projection Operator modifies global spatial organization through coherent perspective deformation. Together, they produce structured transformed views to optimize transferable adversarial perturbations. Experiments on an ImageNet subset show that FRO consistently improves black-box transferability across diverse CNN and Vision Transformer models. We further analyze the effect of implicit ensemble size and evaluate different transformation-based methods under a unified ensemble scale, demonstrating the superiority of designing input transformations from the perspective of front-end response ensembles.

Convolutional Neural Networks and Vision Transformers are the two dominant architectural families in computer vision, defined by spatially local convolution and global self-attention respectively. Despite their apparent differences, we show that both operations are special cases of a single $k$-nearest neighbor aggregation framework: convolution selects neighbors by spatial proximity while attention selects by feature similarity, placing them at two ends of a shared operational spectrum. We introduce Convolutional Nearest Neighbors (ConvNN), a unified framework that exactly recovers standard and depthwise convolution, self-attention, and sparse attention variants including KVT-attention as special cases, and exposes the design space of neighbor-selection strategies between them through configurable similarity functions, positional encodings, and aggregation kernels. We validate ConvNN on ImageNet-1K classification across two complementary architectures: a hybrid branching layer in ResNet-50 that combines local and global feature learning, improving top-1 accuracy by 3.0% over the ResNet-50 baseline, and ConvNN-attention in ViT-Base that achieves 81.64% top-1 accuracy, surpassing standard multi-head self-attention by 0.7%. Together, these results demonstrate that ConvNN provides a principled foundation for designing operations that bridge convolutional and attention-based computation.

Activation-aware low-rank factorization techniques yield strong compression results but are generally confined to linear layers, while existing whitening-based theory typically makes an implicit full-rank assumption on activations. We introduce a layer representation framework that extends activation-aware factorization beyond linear layers, including standard and grouped convolutions. Within this framework, our whitening-based formulation is more general than prior ones, naturally covering rank-deficient activations, and yields an optimal low-rank projection that attains the reconstruction error of the best low-rank approximation to layer activations. The resulting singular spectrum provides a closed-form per-layer distortion proxy, which we use to allocate per-layer ranks under explicit FLOP or parameter-count budgets via a Lagrangian relaxation with negligible overhead. Together, these components form BALF, an end-to-end pipeline for efficient vision model compression. Across CNNs and vision transformers on CIFAR-10 and ImageNet-1K, BALF generally achieves higher accuracy than SVD-based factorization baselines at matched FLOP or parameter count targets and remains competitive with other fine-tuning-free compression techniques.

Human vision is a highly active process driven by gaze, which directs attention to task-relevant regions through foveation, dramatically reducing visual processing. In contrast, robot learning systems typically rely on passive, uniform processing of raw camera images. In this work, we explore how incorporating human-like active gaze into robotic policies can enhance efficiency and robustness. We develop GIAVA (Gaze Integrated Active-Vision ALOHA), a robot vision system that emulates human head and neck movement, and gaze adjustment for foveated processing. Extending the AV-ALOHA robot platform, we introduce a framework for simultaneously collecting eye-tracking, perspective control, and robot manipulation demonstration data from a human operator. We also open-source a simulation benchmark and dataset for training robot policies that incorporate human gaze. Inspired by recent work in foveated image segmentation and given the widespread use of Vision Transformers (ViTs) in robot learning, we integrate gaze information into ViTs using a foveated patch tokenization scheme. Compared to uniform patch tokenization, this significantly reduces the number of tokens, and thus computation. Our results show that our method for foveated robot vision drastically reduces computational overhead, and enhances robustness to background distractors. Notably, on certain high-precision tasks, foveated vision also improves performance, as reflected in higher success rates. Together, these findings suggest that human-inspired foveated visual processing offers untapped potential and should be further considered as a useful inductive bias in robotic vision systems. https://soltanilara.github.io/giava/

We propose a vision transformer (ViT)-based deep learning framework to refine disaster-affected area segmentation from remote sensing imagery, aiming to support and enhance the Emergent Value Added Product (EVAP) developed by the Taiwan Space Agency (TASA). The process starts with a small set of manually annotated regions. We then apply principal component analysis (PCA)-based feature space analysis and construct a confidence index (CI) to expand these labels, producing a weakly supervised training set. These expanded labels are then used to train ViT-based encoder-decoder models with multi-band inputs from Sentinel-2 and Formosat-5 imagery. Our architecture supports multiple decoder variants and multi-stage loss strategies to improve performance under limited supervision. During the evaluation, model predictions are compared with higher-resolution EVAP output to assess spatial coherence and segmentation consistency. Case studies on the 2022 Poyang Lake drought and the 2023 Rhodes wildfire demonstrate that our framework improves the smoothness and reliability of segmentation results, offering a scalable approach for disaster mapping when accurate ground truth is unavailable.

The integration of Large Language Model (LLM) blocks with Vision Transformers (ViTs) holds significant promise for vision-only tasks by leveraging the rich semantic knowledge and reasoning capabilities of LLMs. However, a fundamental challenge lies in the inherent modality mismatch between the text-centric pre-training of LLMs and the vision-centric training of ViTs. Direct fusion often fails to fully exploit the LLM's potential and suffers from unstable finetuning. Consequently, prior works typically keep LLM blocks frozen while learning only the vision components. To address these challenges, we introduce Language-Adapted Vision Enhancer (LAVIE), a novel framework that bridges this modality gap through a synergistic pre-training strategy. LAVIE co-adapts a ViT backbone and an LLM fusion block by (1) employing Masked Auto-Encoding (MAE) to pre-train the ViT for richer visual representations, and (2) concurrently training Low-Rank Adaptation (LoRA) layers within the LLM block using the same MAE objective. This joint optimization guides the ViT to produce LLM-aligned features and the LLM to effectively interpret visual information. We demonstrate through extensive experiments that LAVIE significantly improves performance in various downstream vision tasks, offering an effective and efficient way to enhance visual understanding using frozen LLM knowledge. Code is available at https://github.com/selimkuzucu/LAVIE

Many real-world applications collect data in a streaming environment, where learning tasks are encountered sequentially. This necessitates continual learning (CL) to update models online, enabling adaptation to new tasks while preserving past knowledge to prevent catastrophic forgetting. Nowadays, with the flourish of large pre-trained models (LPMs), efficiency has become increasingly critical for CL, due to their substantial computational demands and growing parameter sizes. In this paper, we introduce TreeLoRA (K-D Tree of Low-Rank Adapters), a novel approach that constructs layer-wise adapters by leveraging hierarchical gradient similarity to enable efficient CL, particularly for LPMs. To reduce the computational burden of task similarity estimation, we employ bandit techniques to develop an algorithm based on lower confidence bounds to efficiently explore the task structure. Furthermore, we use sparse gradient updates to facilitate parameter optimization, making the approach better suited for LPMs. Theoretical analysis is provided to justify the rationale behind our approach, and experiments on both vision transformers (ViTs) and large language models (LLMs) demonstrate the effectiveness and efficiency of our approach across various domains, including vision and natural language processing tasks.

Natural images exhibit strong geometric regularities: local structures, such as edges, corners, and textures, appear in many orientations and mirror configurations. Since Vision Transformers (ViTs) operate on square image patches, these transformations naturally correspond to the dihedral symmetry group $\mathrm{D}_8$, also known as the octic group. Recent work has shown that ViTs can be made reflection equivariant and more efficient than standard ViTs simultaneously by implementing the linear layers in the Fourier domain of the reflection group. In this work, we extend the equivariance to reflections and rotations and analyze the scalability of the resulting networks. Our Quick ViTs, based on octic equivariant linear layers, achieve 5.33x reductions in FLOPs and up to 8x reductions in memory compared to ordinary linear layers. By analyzing the arithmetic intensity of these layers, we identify theoretical limits on how much the FLOP savings translate into throughput improvements on modern GPUs. However, these limitations disappear as the embedding dimensions increase. Enabled by their computational efficiency, we conduct a broader empirical evaluation of equivariant ViTs than in previous work. Upon training supervised (DeiT-III) and self-supervised (DINOv2) on ImageNet-1K, we find that our Quick ViTs match or exceed baseline accuracy while at the same time providing substantial efficiency gains.

Breast cancer survival prediction in computational pathology presents a remarkable challenge due to tumor heterogeneity. For instance, different regions of the same tumor in the pathology image can show distinct morphological and molecular characteristics. This makes it difficult to extract representative features from whole slide images (WSIs) that truly reflect the tumor's aggressive potential and likely survival outcomes. In this paper, we present PathoHR, a novel pipeline for accurate breast cancer survival prediction that enhances any size of pathological images to enable more effective feature learning. Our approach entails (1) the incorporation of a plug-and-play high-resolution Vision Transformer (ViT) to enhance patch-wise WSI representation, enabling more detailed and comprehensive feature extraction, (2) the systematic evaluation of multiple advanced similarity metrics for comparing WSI-extracted features, optimizing the representation learning process to better capture tumor characteristics, (3) the demonstration that smaller image patches enhanced follow the proposed pipeline can achieve equivalent or superior prediction accuracy compared to raw larger patches, while significantly reducing computational overhead. Experimental findings valid that PathoHR provides the potential way of integrating enhanced image resolution with optimized feature learning to advance computational pathology, offering a promising direction for more accurate and efficient breast cancer survival prediction. Code will be available at https://github.com/AIGeeksGroup/PathoHR.

Vision Transformers (ViTs) have achieved strong visual recognition performance, yet their deployment in resource-constrained industrial environments remains limited. The main challenges are their high computational cost, memory requirements, and energy consumption. While individual efficiency techniques such as neural architecture search (NAS), token compression, and low-precision inference have been extensively studied, most prior work targets only a single optimization axis, limiting overall deployment gains while preserving accuracy. In this paper, we present one of the first holistic frameworks that jointly optimizes three complementary axes: architecture, token, and bit-width. Specifically, the framework identifies compact backbones via Neural Architecture Search (AutoFormer), reduces information processing via token merging (ToMe), and accelerates per-operation execution via fp16 mixed-precision inference. In our study, we analyze accuracy-efficiency trade-offs on ImageNet-1K under aggressive compression. We then apply the selected configurations to a real-world in-house 3D X-ray semiconductor defect classification dataset for IC chip packaging inspection. Results show that the proposed multi-axis framework achieves more than $10\times$ improvement in throughput along with over $10\times$ reductions in parameter count, FLOPs, and energy consumption, while maintaining the required accuracy on the downstream industrial task. To the best of our knowledge, this is among the earliest works to jointly optimize architecture, token, and bit-width dimensions in ViTs and the first such resource-efficient, deployment-focused study for semiconductor manufacturing.

Lossy compression is conventionally driven by a task-agnostic distortion (e.g., MSE or MS-SSIM), yet in many emerging applications the receiver cares not about uniform fidelity but about a downstream task whose relevant content varies across the signal and evolves over time. We formulate task-aware compression as a weighted rate-distortion problem, in which a single codec is driven by a separable, per-component weighted distortion whose weights encode task importance and may depend on the source. We introduce task consistency, i.e., that minimizing the weighted distortion also minimizes the true task loss, and characterize when it holds: for linear tasks, the task loss admits a weighted-MSE form with signal-independent weights under suitable cross-term conditions, while for nonlinear tasks, an integrated-gradients analysis motivates separable task-aware weights. We show how task symmetry and irrelevance further constrain the admissible weights. Guided by this theory, we realize the weight-conditioned code in a single learned Vision Transformer (ViT) codec whose token-level attention natively consumes a per-component importance vector, so one fixed backbone is re-targeted at runtime, from universal (task-agnostic) to task-specialized operation, purely by swapping the injected weights, without retraining, while producing a single human-viewable reconstruction steered to the active task. On downstream face-analysis tasks, a single model reaches 91.4% accuracy at 0.034 bpp on a localized task, within 1.9% of a task-specific codec (93.3%) and well above a universal codec (76.9%). Such task-adaptive compression suits bandwidth-constrained perception systems, e.g., in Physical AI, where the active task drifts and per-task retraining is infeasible.

Automatic cardiac image segmentation is pivotal for diagnosing and treating cardiac diseases. In this work, we introduce MCSeg, a volumetric transformer-based network tailored for multi-modal cardiac segmentation. To overcome the architectural mismatch inherent in existing hybrid networks, we propose a novel Scaling Feature Pyramid (SFP). Unlike conventional skip connections, the SFP effectively bridges the single-scale 3D Vision Transformer (ViT) encoder and the multi-scale CNN decoder by transforming the ViT's output into a hierarchical feature pyramid, ensuring that global contextual information is effectively leveraged. For the training paradigm, the ViT encoder first undergoes self-supervised pre-training via masked image modeling. Subsequently, the network is fine-tuned on downstream tasks, during which a regional mutual information (RMI) loss is integrated to improve boundary segmentation accuracy. In experiments, MCSeg consistently outperforms eleven SOTA methods on CT dataset ImageCHD, multi-modal dataset MM-WHS, MRI dataset HVSMR-2.0 and MSD Heart, highlighting the effectiveness of our MCSeg for multi-modal cardiac segmentation tasks. Furthermore, MCSeg's superior performance in few-shot experiment showcases its significant potential in adapting to limited data scenarios. Codes and pre-trained ViT-B weights are open-sourced at https://openi.pcl.ac.cn/OpenMedIA/MCSeg