PyTorchKR Papers

최신 논문

총 4,526편
태그·기간 등 상세 필터

카테고리 탐색

로그인 후 이 태그 구독

arXiv:2403.07292공식 코드 · Papers with Code

Continual All-in-One Adverse Weather Removal with Knowledge Replay on a Unified Network Structure

In real-world applications, image degeneration caused by adverse weather is always complex and changes with different weather conditions from days and seasons. Systems in real-world environments constantly encounter adverse weather conditions that are not previously observed. Therefore, it practically requires adverse weather removal models to continually learn from incrementally collected data reflecting various degeneration types. Existing adverse weather removal approaches, for either single or multiple adverse weathers, are mainly designed for a static learning paradigm, which assumes that the data of all types of degenerations to handle can be finely collected at one time before a single-phase learning process. They thus cannot directly handle the incremental learning requirements. To address this issue, we made the earliest effort to investigate the continual all-in-one adverse weather removal task, in a setting closer to real-world applications. Specifically, we develop a novel continual learning framework with effective knowledge replay (KR) on a unified network structure. Equipped with a principal component projection and an effective knowledge distillation mechanism, the proposed KR techniques are tailored for the all-in-one weather removal task. It considers the characteristics of the image restoration task with multiple degenerations in continual learning, and the knowledge for different degenerations can be shared and accumulated in the unified network structure. Extensive experimental results demonstrate the effectiveness of the proposed method to deal with this challenging task, which performs competitively to existing dedicated or joint training image restoration methods. Our code is available at https://github.com/xiaojihh/CL_all-in-one.

arXiv:2403.06414

Evolving Knowledge Distillation with Large Language Models and Active Learning

Large language models (LLMs) have demonstrated remarkable capabilities across various NLP tasks. However, their computational costs are prohibitively high. To address this issue, previous research has attempted to distill the knowledge of LLMs into smaller models by generating annotated data. Nonetheless, these works have mainly focused on the direct use of LLMs for text generation and labeling, without fully exploring their potential to comprehend the target task and acquire valuable knowledge. In this paper, we propose EvoKD: Evolving Knowledge Distillation, which leverages the concept of active learning to interactively enhance the process of data generation using large language models, simultaneously improving the task capabilities of small domain model (student model). Different from previous work, we actively analyze the student model's weaknesses, and then synthesize labeled samples based on the analysis. In addition, we provide iterative feedback to the LLMs regarding the student model's performance to continuously construct diversified and challenging samples. Experiments and analysis on different NLP tasks, namely, text classification and named entity recognition show the effectiveness of EvoKD.

arXiv:2403.06089

Knowledge Distillation of Convolutional Neural Networks through Feature Map Transformation using Decision Trees

The interpretation of reasoning by Deep Neural Networks (DNN) is still challenging due to their perceived black-box nature. Therefore, deploying DNNs in several real-world tasks is restricted by the lack of transparency of these models. We propose a distillation approach by extracting features from the final layer of the convolutional neural network (CNN) to address insights to its reasoning. The feature maps in the final layer of a CNN are transformed into a one-dimensional feature vector using a fully connected layer. Subsequently, the extracted features are used to train a decision tree to achieve the best accuracy under constraints of depth and nodes. We use the medical images of dermaMNIST, octMNIST, and pneumoniaMNIST from the medical MNIST datasets to demonstrate our proposed work. We observed that performance of the decision tree is as good as a CNN with minimum complexity. The results encourage interpreting decisions made by the CNNs using decision trees.

arXiv:2403.06071공식 코드 · Papers with Code

Bit-mask Robust Contrastive Knowledge Distillation for Unsupervised Semantic Hashing

Unsupervised semantic hashing has emerged as an indispensable technique for fast image search, which aims to convert images into binary hash codes without relying on labels. Recent advancements in the field demonstrate that employing large-scale backbones (e.g., ViT) in unsupervised semantic hashing models can yield substantial improvements. However, the inference delay has become increasingly difficult to overlook. Knowledge distillation provides a means for practical model compression to alleviate this delay. Nevertheless, the prevailing knowledge distillation approaches are not explicitly designed for semantic hashing. They ignore the unique search paradigm of semantic hashing, the inherent necessities of the distillation process, and the property of hash codes. In this paper, we propose an innovative Bit-mask Robust Contrastive knowledge Distillation (BRCD) method, specifically devised for the distillation of semantic hashing models. To ensure the effectiveness of two kinds of search paradigms in the context of semantic hashing, BRCD first aligns the semantic spaces between the teacher and student models through a contrastive knowledge distillation objective. Additionally, to eliminate noisy augmentations and ensure robust optimization, a cluster-based method within the knowledge distillation process is introduced. Furthermore, through a bit-level analysis, we uncover the presence of redundancy bits resulting from the bit independence property. To mitigate these effects, we introduce a bit mask mechanism in our knowledge distillation objective. Finally, extensive experiments not only showcase the noteworthy performance of our BRCD method in comparison to other knowledge distillation methods but also substantiate the generality of our methods across diverse semantic hashing models and backbones. The code for BRCD is available at https://github.com/hly1998/BRCD.

arXiv:2403.05894공식 코드 · Papers with Code

Frequency Attention for Knowledge Distillation

Knowledge distillation is an attractive approach for learning compact deep neural networks, which learns a lightweight student model by distilling knowledge from a complex teacher model. Attention-based knowledge distillation is a specific form of intermediate feature-based knowledge distillation that uses attention mechanisms to encourage the student to better mimic the teacher. However, most of the previous attention-based distillation approaches perform attention in the spatial domain, which primarily affects local regions in the input image. This may not be sufficient when we need to capture the broader context or global information necessary for effective knowledge transfer. In frequency domain, since each frequency is determined from all pixels of the image in spatial domain, it can contain global information about the image. Inspired by the benefits of the frequency domain, we propose a novel module that functions as an attention mechanism in the frequency domain. The module consists of a learnable global filter that can adjust the frequencies of student's features under the guidance of the teacher's features, which encourages the student's features to have patterns similar to the teacher's features. We then propose an enhanced knowledge review-based distillation model by leveraging the proposed frequency attention module. The extensive experiments with various teacher and student architectures on image classification and object detection benchmark datasets show that the proposed approach outperforms other knowledge distillation methods.

arXiv:2403.05796공식 코드 · Papers with Code

Weakly Supervised Change Detection via Knowledge Distillation and Multiscale Sigmoid Inference

Change detection, which aims to detect spatial changes from a pair of multi-temporal images due to natural or man-made causes, has been widely applied in remote sensing, disaster management, urban management, etc. Most existing change detection approaches, however, are fully supervised and require labor-intensive pixel-level labels. To address this, we develop a novel weakly supervised change detection technique via Knowledge Distillation and Multiscale Sigmoid Inference (KD-MSI) that leverages image-level labels. In our approach, the Class Activation Maps (CAM) are utilized not only to derive a change probability map but also to serve as a foundation for the knowledge distillation process. This is done through a joint training strategy of the teacher and student networks, enabling the student network to highlight potential change areas more accurately than teacher network based on image-level labels. Moreover, we designed a Multiscale Sigmoid Inference (MSI) module as a post processing step to further refine the change probability map from the trained student network. Empirical results on three public datasets, i.e., WHU-CD, DSIFN-CD, and LEVIR-CD, demonstrate that our proposed technique, with its integrated training strategy, significantly outperforms the state-of-the-art.

arXiv:2403.05687

Scene Graph Aided Radiology Report Generation

Radiology report generation (RRG) methods often lack sufficient medical knowledge to produce clinically accurate reports. The scene graph contains rich information to describe the objects in an image. We explore enriching the medical knowledge for RRG via a scene graph, which has not been done in the current RRG literature. To this end, we propose the Scene Graph aided RRG (SGRRG) network, a framework that generates region-level visual features, predicts anatomical attributes, and leverages an automatically generated scene graph, thus achieving medical knowledge distillation in an end-to-end manner. SGRRG is composed of a dedicated scene graph encoder responsible for translating the scene graph, and a scene graph-aided decoder that takes advantage of both patch-level and region-level visual information. A fine-grained, sentence-level attention method is designed to better dis-till the scene graph information. Extensive experiments demonstrate that SGRRG outperforms previous state-of-the-art methods in report generation and can better capture abnormal findings.

arXiv:2403.06679공식 코드 · Papers with Code

Answering Diverse Questions via Text Attached with Key Audio-Visual Clues

Audio-visual question answering (AVQA) requires reference to video content and auditory information, followed by correlating the question to predict the most precise answer. Although mining deeper layers of audio-visual information to interact with questions facilitates the multimodal fusion process, the redundancy of audio-visual parameters tends to reduce the generalization of the inference engine to multiple question-answer pairs in a single video. Indeed, the natural heterogeneous relationship between audiovisuals and text makes the perfect fusion challenging, to prevent high-level audio-visual semantics from weakening the network's adaptability to diverse question types, we propose a framework for performing mutual correlation distillation (MCD) to aid question inference. MCD is divided into three main steps: 1) firstly, the residual structure is utilized to enhance the audio-visual soft associations based on self-attention, then key local audio-visual features relevant to the question context are captured hierarchically by shared aggregators and coupled in the form of clues with specific question vectors. 2) Secondly, knowledge distillation is enforced to align audio-visual-text pairs in a shared latent space to narrow the cross-modal semantic gap. 3) And finally, the audio-visual dependencies are decoupled by discarding the decision-level integrations. We evaluate the proposed method on two publicly available datasets containing multiple question-and-answer pairs, i.e., Music-AVQA and AVQA. Experiments show that our method outperforms other state-of-the-art methods, and one interesting finding behind is that removing deep audio-visual features during inference can effectively mitigate overfitting. The source code is released at http://github.com/rikeilong/MCD-forAVQA.

arXiv:2403.06417

Enhanced Sparsification via Stimulative Training

Sparsification-based pruning has been an important category in model compression. Existing methods commonly set sparsity-inducing penalty terms to suppress the importance of dropped weights, which is regarded as the suppressed sparsification paradigm. However, this paradigm inactivates the dropped parts of networks causing capacity damage before pruning, thereby leading to performance degradation. To alleviate this issue, we first study and reveal the relative sparsity effect in emerging stimulative training and then propose a structured pruning framework, named STP, based on an enhanced sparsification paradigm which maintains the magnitude of dropped weights and enhances the expressivity of kept weights by self-distillation. Besides, to find an optimal architecture for the pruned network, we propose a multi-dimension architecture space and a knowledge distillation-guided exploration strategy. To reduce the huge capacity gap of distillation, we propose a subnet mutating expansion technique. Extensive experiments on various benchmarks indicate the effectiveness of STP. Specifically, without fine-tuning, our method consistently achieves superior performance at different budgets, especially under extremely aggressive pruning scenarios, e.g., remaining 95.11% Top-1 accuracy (72.43% in 76.15%) while reducing 85% FLOPs for ResNet-50 on ImageNet. Codes will be released soon.

arXiv:2403.06239공식 코드 · Papers with Code

Cooperative Classification and Rationalization for Graph Generalization

Graph Neural Networks (GNNs) have achieved impressive results in graph classification tasks, but they struggle to generalize effectively when faced with out-of-distribution (OOD) data. Several approaches have been proposed to address this problem. Among them, one solution is to diversify training distributions in vanilla classification by modifying the data environment, yet accessing the environment information is complex. Besides, another promising approach involves rationalization, extracting invariant rationales for predictions. However, extracting rationales is difficult due to limited learning signals, resulting in less accurate rationales and diminished predictions. To address these challenges, in this paper, we propose a Cooperative Classification and Rationalization (C2R) method, consisting of the classification and the rationalization module. Specifically, we first assume that multiple environments are available in the classification module. Then, we introduce diverse training distributions using an environment-conditional generative network, enabling robust graph representations. Meanwhile, the rationalization module employs a separator to identify relevant rationale subgraphs while the remaining non-rationale subgraphs are de-correlated with labels. Next, we align graph representations from the classification module with rationale subgraph representations using the knowledge distillation methods, enhancing the learning signal for rationales. Finally, we infer multiple environments by gathering non-rationale representations and incorporate them into the classification module for cooperative learning. Extensive experimental results on both benchmarks and synthetic datasets demonstrate the effectiveness of C2R. Code is available at https://github.com/yuelinan/Codes-of-C2R.

arXiv:2403.06213공식 코드 · Papers with Code

$V_kD:$ Improving Knowledge Distillation using Orthogonal Projections

Knowledge distillation is an effective method for training small and efficient deep learning models. However, the efficacy of a single method can degenerate when transferring to other tasks, modalities, or even other architectures. To address this limitation, we propose a novel constrained feature distillation method. This method is derived from a small set of core principles, which results in two emerging components: an orthogonal projection and a task-specific normalisation. Equipped with both of these components, our transformer models can outperform all previous methods on ImageNet and reach up to a 4.4% relative improvement over the previous state-of-the-art methods. To further demonstrate the generality of our method, we apply it to object detection and image generation, whereby we obtain consistent and substantial performance improvements over state-of-the-art. Code and models are publicly available: https://github.com/roymiles/vkd

arXiv:2403.06914공식 코드 · Papers with Code

MEND: Meta dEmonstratioN Distillation for Efficient and Effective In-Context Learning

Large Language models (LLMs) have demonstrated impressive in-context learning (ICL) capabilities, where a LLM makes predictions for a given test input together with a few input-output pairs (demonstrations). Nevertheless, the inclusion of demonstrations leads to a quadratic increase in the computational overhead of the self-attention mechanism. Existing solutions attempt to distill lengthy demonstrations into compact vectors. However, they often require task-specific retraining or compromise LLM's in-context learning performance. To mitigate these challenges, we present Meta dEmonstratioN Distillation (MEND), where a language model learns to distill any lengthy demonstrations into vectors without retraining for a new downstream task. We exploit the knowledge distillation to enhance alignment between MEND and LLM, achieving both efficiency and effectiveness simultaneously. MEND is endowed with the meta-knowledge of distilling demonstrations through a two-stage training process, which includes meta-distillation pretraining and fine-tuning. Comprehensive evaluations across seven diverse ICL task partitions using decoder-only (GPT-2) and encoder-decoder (T5) attest to MEND's prowess. It not only matches but often outperforms the Vanilla ICL as well as other state-of-the-art distillation models, while significantly reducing the computational demands. This innovation promises enhanced scalability and efficiency for the practical deployment of large language models

arXiv:2403.05451

Attention-guided Feature Distillation for Semantic Segmentation

In contrast to existing complex methodologies commonly employed for distilling knowledge from a teacher to a student, this paper showcases the efficacy of a simple yet powerful method for utilizing refined feature maps to transfer attention. The proposed method has proven to be effective in distilling rich information, outperforming existing methods in semantic segmentation as a dense prediction task. The proposed Attention-guided Feature Distillation (AttnFD) method, employs the Convolutional Block Attention Module (CBAM), which refines feature maps by taking into account both channel-specific and spatial information content. Simply using the Mean Squared Error (MSE) loss function between the refined feature maps of the teacher and the student, AttnFD demonstrates outstanding performance in semantic segmentation, achieving state-of-the-art results in terms of improving the mean Intersection over Union (mIoU) of the student network on the PascalVoc 2012, Cityscapes, COCO, and CamVid datasets.

arXiv:2403.05325

Fine-tuning a Multiple Instance Learning Feature Extractor with Masked Context Modelling and Knowledge Distillation

The first step in Multiple Instance Learning (MIL) algorithms for Whole Slide Image (WSI) classification consists of tiling the input image into smaller patches and computing their feature vectors produced by a pre-trained feature extractor model. Feature extractor models that were pre-trained with supervision on ImageNet have proven to transfer well to this domain, however, this pre-training task does not take into account that visual information in neighboring patches is highly correlated. Based on this observation, we propose to increase downstream MIL classification by fine-tuning the feature extractor model using \textit{Masked Context Modelling with Knowledge Distillation}. In this task, the feature extractor model is fine-tuned by predicting masked patches in a bigger context window. Since reconstructing the input image would require a powerful image generation model, and our goal is not to generate realistically looking image patches, we predict instead the feature vectors produced by a larger teacher network. A single epoch of the proposed task suffices to increase the downstream performance of the feature-extractor model when used in a MIL scenario, even capable of outperforming the downstream performance of the teacher model, while being considerably smaller and requiring a fraction of its compute.

arXiv:2403.05181

Adversarial Sparse Teacher: Defense Against Distillation-Based Model Stealing Attacks Using Adversarial Examples

We introduce Adversarial Sparse Teacher (AST), a robust defense method against distillation-based model stealing attacks. Our approach trains a teacher model using adversarial examples to produce sparse logit responses and increase the entropy of the output distribution. Typically, a model generates a peak in its output corresponding to its prediction. By leveraging adversarial examples, AST modifies the teacher model's original response, embedding a few altered logits into the output while keeping the primary response slightly higher. Concurrently, all remaining logits are elevated to further increase the output distribution's entropy. All these complex manipulations are performed using an optimization function with our proposed Exponential Predictive Divergence (EPD) loss function. EPD allows us to maintain higher entropy levels compared to traditional KL divergence, effectively confusing attackers. Experiments on CIFAR-10 and CIFAR-100 datasets demonstrate that AST outperforms state-of-the-art methods, providing effective defense against model stealing while preserving high accuracy. The source codes will be made publicly available here soon.

arXiv:2403.05061공식 코드 · Papers with Code

RadarDistill: Boosting Radar-based Object Detection Performance via Knowledge Distillation from LiDAR Features

The inherent noisy and sparse characteristics of radar data pose challenges in finding effective representations for 3D object detection. In this paper, we propose RadarDistill, a novel knowledge distillation (KD) method, which can improve the representation of radar data by leveraging LiDAR data. RadarDistill successfully transfers desirable characteristics of LiDAR features into radar features using three key components: Cross-Modality Alignment (CMA), Activation-based Feature Distillation (AFD), and Proposal-based Feature Distillation (PFD). CMA enhances the density of radar features by employing multiple layers of dilation operations, effectively addressing the challenge of inefficient knowledge transfer from LiDAR to radar. AFD selectively transfers knowledge based on regions of the LiDAR features, with a specific focus on areas where activation intensity exceeds a predefined threshold. PFD similarly guides the radar network to selectively mimic features from the LiDAR network within the object proposals. Our comparative analyses conducted on the nuScenes datasets demonstrate that RadarDistill achieves state-of-the-art (SOTA) performance for radar-only object detection task, recording 20.5% in mAP and 43.7% in NDS. Also, RadarDistill significantly improves the performance of the camera-radar fusion model.

arXiv:2403.04908공식 코드 · Papers with Code

Self-Adapting Large Visual-Language Models to Edge Devices across Visual Modalities

Recent advancements in Vision-Language (VL) models have sparked interest in their deployment on edge devices, yet challenges in handling diverse visual modalities, manual annotation, and computational constraints remain. We introduce EdgeVL, a novel framework that bridges this gap by seamlessly integrating dual-modality knowledge distillation and quantization-aware contrastive learning. This approach enables the adaptation of large VL models, like CLIP, for efficient use with both RGB and non-RGB images on resource-limited devices without the need for manual annotations. EdgeVL not only transfers visual language alignment capabilities to compact models but also maintains feature quality post-quantization, significantly enhancing open-vocabulary classification performance across various visual modalities. Our work represents the first systematic effort to adapt large VL models for edge deployment, showcasing up to 15.4% accuracy improvements on multiple datasets and up to 93-fold reduction in model size.

논문 ID: document/1043821

Distilled Gradual Pruning with Pruned Fine-tuning

Neural Networks (NNs) have been driving machine learning progress in recent years, but their larger models present challenges in resource-limited environments. Weight pruning reduces the computational demand, often with performance degradation and long training procedures. This work introduces Distilled Gradual Pruning with Pruned Fine-tuning (DG2PF), a comprehensive algorithm that iteratively prunes pre-trained neural networks using knowledge distillation. We employ a magnitude-based unstructured pruning function that selectively removes a specified proportion of unimportant weights from the network. This function also leads to an efficient compression of the model size while minimizing classification accuracy loss. Additionally, we introduce a simulated pruning strategy with the same effects of weight recovery but while maintaining stable convergence. Furthermore, we propose a multi-step self-knowledge distillation strategy to effectively transfer the knowledge of the full, unpruned network to the pruned counterpart. We validate the performance of our algorithm through extensive experimentation on diverse benchmark datasets, including CIFAR-10 and ImageNet, as well as a set of model architectures. The results highlight how our algorithm prunes and optimizes pretrained neural networks without substantially degrading their classification accuracy while delivering significantly faster and more compact models.

arXiv:2403.04260

Can Small Language Models be Good Reasoners for Sequential Recommendation?

Large language models (LLMs) open up new horizons for sequential recommendations, owing to their remarkable language comprehension and generation capabilities. However, there are still numerous challenges that should be addressed to successfully implement sequential recommendations empowered by LLMs. Firstly, user behavior patterns are often complex, and relying solely on one-step reasoning from LLMs may lead to incorrect or task-irrelevant responses. Secondly, the prohibitively resource requirements of LLM (e.g., ChatGPT-175B) are overwhelmingly high and impractical for real sequential recommender systems. In this paper, we propose a novel Step-by-step knowLedge dIstillation fraMework for recommendation (SLIM), paving a promising path for sequential recommenders to enjoy the exceptional reasoning capabilities of LLMs in a "slim" (i.e., resource-efficient) manner. We introduce CoT prompting based on user behavior sequences for the larger teacher model. The rationales generated by the teacher model are then utilized as labels to distill the downstream smaller student model (e.g., LLaMA2-7B). In this way, the student model acquires the step-by-step reasoning capabilities in recommendation tasks. We encode the generated rationales from the student model into a dense vector, which empowers recommendation in both ID-based and ID-agnostic scenarios. Extensive experiments demonstrate the effectiveness of SLIM over state-of-the-art baselines, and further analysis showcasing its ability to generate meaningful recommendation reasoning at affordable costs.

arXiv:2403.04245공식 코드 · Papers with Code

A Study of Dropout-Induced Modality Bias on Robustness to Missing Video Frames for Audio-Visual Speech Recognition

Advanced Audio-Visual Speech Recognition (AVSR) systems have been observed to be sensitive to missing video frames, performing even worse than single-modality models. While applying the dropout technique to the video modality enhances robustness to missing frames, it simultaneously results in a performance loss when dealing with complete data input. In this paper, we investigate this contrasting phenomenon from the perspective of modality bias and reveal that an excessive modality bias on the audio caused by dropout is the underlying reason. Moreover, we present the Modality Bias Hypothesis (MBH) to systematically describe the relationship between modality bias and robustness against missing modality in multimodal systems. Building on these findings, we propose a novel Multimodal Distribution Approximation with Knowledge Distillation (MDA-KD) framework to reduce over-reliance on the audio modality and to maintain performance and robustness simultaneously. Finally, to address an entirely missing modality, we adopt adapters to dynamically switch decision strategies. The effectiveness of our proposed approach is evaluated and validated through a series of comprehensive experiments using the MISP2021 and MISP2022 datasets. Our code is available at https://github.com/dalision/ModalBiasAVSR