PyTorchKR Papers

최신 논문

총 232편
태그·기간 등 상세 필터

카테고리 탐색

로그인 후 이 태그 구독

arXiv:2405.16148공식 코드 · Papers with Code

Accelerating Transformers with Spectrum-Preserving Token Merging

Increasing the throughput of the Transformer architecture, a foundational component used in numerous state-of-the-art models for vision and language tasks (e.g., GPT, LLaVa), is an important problem in machine learning. One recent and effective strategy is to merge token representations within Transformer models, aiming to reduce computational and memory requirements while maintaining accuracy. Prior works have proposed algorithms based on Bipartite Soft Matching (BSM), which divides tokens into distinct sets and merges the top k similar tokens. However, these methods have significant drawbacks, such as sensitivity to token-splitting strategies and damage to informative tokens in later layers. This paper presents a novel paradigm called PiToMe, which prioritizes the preservation of informative tokens using an additional metric termed the energy score. This score identifies large clusters of similar tokens as high-energy, indicating potential candidates for merging, while smaller (unique and isolated) clusters are considered as low-energy and preserved. Experimental findings demonstrate that PiToMe saved from 40-60\% FLOPs of the base models while exhibiting superior off-the-shelf performance on image classification (0.5\% average performance drop of ViT-MAE-H compared to 2.6\% as baselines), image-text retrieval (0.3\% average performance drop of CLIP on Flickr30k compared to 4.5\% as others), and analogously in visual questions answering with LLaVa-7B. Furthermore, PiToMe is theoretically shown to preserve intrinsic spectral properties of the original token space under mild conditions

arXiv:2410.15266공식 코드 · Papers with Code

GSSF: Generalized Structural Sparse Function for Deep Cross-modal Metric Learning

Cross-modal metric learning is a prominent research topic that bridges the semantic heterogeneity between vision and language. Existing methods frequently utilize simple cosine or complex distance metrics to transform the pairwise features into a similarity score, which suffers from an inadequate or inefficient capability for distance measurements. Consequently, we propose a Generalized Structural Sparse Function to dynamically capture thorough and powerful relationships across modalities for pair-wise similarity learning while remaining concise but efficient. Specifically, the distance metric delicately encapsulates two formats of diagonal and block-diagonal terms, automatically distinguishing and highlighting the cross-channel relevancy and dependency inside a structured and organized topology. Hence, it thereby empowers itself to adapt to the optimal matching patterns between the paired features and reaches a sweet spot between model complexity and capability. Extensive experiments on cross-modal and two extra uni-modal retrieval tasks (image-text retrieval, person re-identification, fine-grained image retrieval) have validated its superiority and flexibility over various popular retrieval frameworks. More importantly, we further discover that it can be seamlessly incorporated into multiple application scenarios, and demonstrates promising prospects from Attention Mechanism to Knowledge Distillation in a plug-and-play manner. Our code is publicly available at: https://github.com/Paranioar/GSSF.

arXiv:2410.11963

CtrlSynth: Controllable Image Text Synthesis for Data-Efficient Multimodal Learning

Pretraining robust vision or multimodal foundation models (e.g., CLIP) relies on large-scale datasets that may be noisy, potentially misaligned, and have long-tail distributions. Previous works have shown promising results in augmenting datasets by generating synthetic samples. However, they only support domain-specific ad hoc use cases (e.g., either image or text only, but not both), and are limited in data diversity due to a lack of fine-grained control over the synthesis process. In this paper, we design a \emph{controllable} image-text synthesis pipeline, CtrlSynth, for data-efficient and robust multimodal learning. The key idea is to decompose the visual semantics of an image into basic elements, apply user-specified control policies (e.g., remove, add, or replace operations), and recompose them to synthesize images or texts. The decompose and recompose feature in CtrlSynth allows users to control data synthesis in a fine-grained manner by defining customized control policies to manipulate the basic elements. CtrlSynth leverages the capabilities of pretrained foundation models such as large language models or diffusion models to reason and recompose basic elements such that synthetic samples are natural and composed in diverse ways. CtrlSynth is a closed-loop, training-free, and modular framework, making it easy to support different pretrained models. With extensive experiments on 31 datasets spanning different vision and vision-language tasks, we show that CtrlSynth substantially improves zero-shot classification, image-text retrieval, and compositional reasoning performance of CLIP models.

arXiv:2410.05346

AnyAttack: Towards Large-scale Self-supervised Adversarial Attacks on Vision-language Models

Due to their multimodal capabilities, Vision-Language Models (VLMs) have found numerous impactful applications in real-world scenarios. However, recent studies have revealed that VLMs are vulnerable to image-based adversarial attacks. Traditional targeted adversarial attacks require specific targets and labels, limiting their real-world impact.We present AnyAttack, a self-supervised framework that transcends the limitations of conventional attacks through a novel foundation model approach. By pre-training on the massive LAION-400M dataset without label supervision, AnyAttack achieves unprecedented flexibility - enabling any image to be transformed into an attack vector targeting any desired output across different VLMs.This approach fundamentally changes the threat landscape, making adversarial capabilities accessible at an unprecedented scale. Our extensive validation across five open-source VLMs (CLIP, BLIP, BLIP2, InstructBLIP, and MiniGPT-4) demonstrates AnyAttack's effectiveness across diverse multimodal tasks. Most concerning, AnyAttack seamlessly transfers to commercial systems including Google Gemini, Claude Sonnet, Microsoft Copilot and OpenAI GPT, revealing a systemic vulnerability requiring immediate attention.

arXiv:2409.19425

From Unimodal to Multimodal: Scaling up Projectors to Align Modalities

Recent contrastive multimodal vision-language models like CLIP have demonstrated robust open-world semantic understanding, becoming the standard image backbones for vision-language applications due to their aligned latent space. However, this practice has left powerful unimodal encoders for both vision and language underutilized in multimodal applications which raises a key question: Is there a plausible way to connect unimodal backbones for zero-shot vision-language tasks? To this end, we propose a novel approach that aligns vision and language modalities using only projection layers on pretrained, frozen unimodal encoders. Our method exploits the high semantic similarity between embedding spaces of well-trained vision and language models. It involves selecting semantically similar encoders in the latent space, curating a concept-rich dataset of image-caption pairs, and training simple MLP projectors. We evaluated our approach on 12 zero-shot classification datasets and 2 image-text retrieval datasets. Our best model, utilizing DINOv2 and All-Roberta-Large text encoder, achieves 76\(\%\) accuracy on ImageNet with a 20-fold reduction in data and 65 fold reduction in compute requirements. The proposed framework enhances the accessibility of model development while enabling flexible adaptation across diverse scenarios, offering an efficient approach to building multimodal models by utilizing existing unimodal architectures. Code and datasets will be released soon.

arXiv:2409.09582

NEVLP: Noise-Robust Framework for Efficient Vision-Language Pre-training

The success of Vision Language Models (VLMs) on various vision-language tasks heavily relies on pre-training with large scale web-crawled datasets. However, the noisy and incomplete nature of web data makes dataset scale crucial for performance, rendering end-to-end training increasingly prohibitive. In this paper, we propose NEVLP, a noise-robust framework for efficient vision-language pre-training that requires less pre-training data. Specifically, we bridge the modality gap between a frozen image encoder and a large language model with a transformer and introduce two innovative learning strategies: noise-adaptive learning and concept-enhanced learning to mitigate the impact of noise. In noise-adaptive learning, we estimate the noise probability of each image-text pair based on the transformer's memorization effect and employ noise-adaptive regularization on image-text contrastive learning to condition cross-modal alignment. In concept-enhanced learning, we enrich incomplete text by incorporating visual concepts (objects in the image) to provide prior information about existing objects for image-text matching and image-grounded text generation, thereby mitigating text incompletion. Our framework effectively utilizes noisy web data and achieves state-of-the-art performance with less pre-training data across a wide range of vision-language tasks, including image-text retrieval, image captioning, and visual question answering.

arXiv:2409.07048

Pushing the Limits of Vision-Language Models in Remote Sensing without Human Annotations

The prominence of generalized foundation models in vision-language integration has witnessed a surge, given their multifarious applications. Within the natural domain, the procurement of vision-language datasets to construct these foundation models is facilitated by their abundant availability and the ease of web crawling. Conversely, in the remote sensing domain, although vision-language datasets exist, their volume is suboptimal for constructing robust foundation models. This study introduces an approach to curate vision-language datasets by employing an image decoding machine learning model, negating the need for human-annotated labels. Utilizing this methodology, we amassed approximately 9.6 million vision-language paired datasets in VHR imagery. The resultant model outperformed counterparts that did not leverage publicly available vision-language datasets, particularly in downstream tasks such as zero-shot classification, semantic localization, and image-text retrieval. Moreover, in tasks exclusively employing vision encoders, such as linear probing and k-NN classification, our model demonstrated superior efficacy compared to those relying on domain-specific vision-language datasets.

논문 ID: document/1053028

Global–Local Information Soft-Alignment for Cross-Modal Remote-Sensing Image–Text Retrieval

Cross-modal remote-sensing image–text retrieval (CMRSITR) is a challenging task that aims to retrieve target remote-sensing (RS) images based on textual descriptions. However, the modal gap between texts and RS images poses a significant challenge. RS images comprise multiple targets and complex backgrounds, necessitating the mining of both global and local information (GaLR) for effective CMRSITR. Existing approaches primarily focus on local image features while disregarding the local features of the text and their correspondence. These methods typically fuse global and local image features and align them with global text features. However, they struggle to eliminate the influence of cluttered backgrounds and may overlook crucial targets. To address these limitations, we propose a novel framework for CMRSITR based on a transformer architecture, which leverages global–local information soft alignment (GLISA) to enhance retrieval performance. Our framework incorporates a global image extraction module, which captures the global semantic features of image–text pairs and effectively represents the relationships among multiple targets in RS images. In addition, we introduce an adaptive local information extraction (ALIE) module that adaptively mines discriminative local clues from both RS images and texts, aligning the corresponding fine-grained information. To mitigate semantic ambiguities during the alignment of local features, we design a local information soft-alignment (LISA) module. In comparative evaluations using two public CMRSITR datasets, our proposed method achieves state-of-the-art results, surpassing not only traditional cross-modal retrieval methods by a substantial margin but also other contrastive language-image pretraining (CLIP)-based methods.

논문 ID: 6772.36578

UGNCL: Uncertainty-Guided Noisy Correspondence Learning for Efficient Cross-Modal Matching

Cross-modal matching has recently gained significant popularity to facilitate retrieval across multi-modal data, and existing works are highly relied on an implicit assumption that the training data pairs are perfectly aligned. However, such an ideal assumption is extremely impossible due to the inevitably mismatched data pairs, a.k.a. noisy correspondence, which can wrongly enforce the mismatched data to be similar and thus induces the performance degradation. Although some recent methods have attempted to address this problem, they still face two challenging issues: 1) un- reliable data division for training inefficiency and 2) unstable pre- diction for matching failure. To address these problems, we pro- pose an efficient Uncertainty-Guided Noisy Correspondence Learning (UGNCL) framework to achieve noise-robust cross-modal matching. Specifically, a novel Uncertainty Guided Division (UGD) algorithm is reliably designed leverage the potential benefits of derived un- certainty to divide the data into clean, noisy and hard partitions, which can effortlessly mitigate the impact of easily-determined noisy pairs. Meanwhile, an efficient Trusted Robust Loss (TRL) is explicitly designed to recast the soft margins, calibrated by confi- dent yet error soft correspondence labels, for the data pairs in the hard partition through the uncertainty, leading to increase/decrease the importance of matched/mismatched pairs and further alleviate the impact of noisy pairs for robustness improvement. Extensive experiments conducted on three public datasets highlight the su- periorities of the proposed framework, and show its competitive performance compared with the state-of-the-arts. The code is avail- able at https://github.com/qxzha/UGNCL.

arXiv:2408.01363

Toward Automatic Relevance Judgment using Vision--Language Models for Image--Text Retrieval Evaluation

Vision--Language Models (VLMs) have demonstrated success across diverse applications, yet their potential to assist in relevance judgments remains uncertain. This paper assesses the relevance estimation capabilities of VLMs, including CLIP, LLaVA, and GPT-4V, within a large-scale \textit{ad hoc} retrieval task tailored for multimedia content creation in a zero-shot fashion. Preliminary experiments reveal the following: (1) Both LLaVA and GPT-4V, encompassing open-source and closed-source visual-instruction-tuned Large Language Models (LLMs), achieve notable Kendall's $\tau \sim 0.4$ when compared to human relevance judgments, surpassing the CLIPScore metric. (2) While CLIPScore is strongly preferred, LLMs are less biased towards CLIP-based retrieval systems. (3) GPT-4V's score distribution aligns more closely with human judgments than other models, achieving a Cohen's $\kappa$ value of around 0.08, which outperforms CLIPScore at approximately -0.096. These findings underscore the potential of LLM-powered VLMs in enhancing relevance judgments.

arXiv:2408.01349공식 코드 · Papers with Code

PC$^2$: Pseudo-Classification Based Pseudo-Captioning for Noisy Correspondence Learning in Cross-Modal Retrieval

In the realm of cross-modal retrieval, seamlessly integrating diverse modalities within multimedia remains a formidable challenge, especially given the complexities introduced by noisy correspondence learning (NCL). Such noise often stems from mismatched data pairs, which is a significant obstacle distinct from traditional noisy labels. This paper introduces Pseudo-Classification based Pseudo-Captioning (PC$^2$) framework to address this challenge. PC$^2$ offers a threefold strategy: firstly, it establishes an auxiliary "pseudo-classification" task that interprets captions as categorical labels, steering the model to learn image-text semantic similarity through a non-contrastive mechanism. Secondly, unlike prevailing margin-based techniques, capitalizing on PC$^2$'s pseudo-classification capability, we generate pseudo-captions to provide more informative and tangible supervision for each mismatched pair. Thirdly, the oscillation of pseudo-classification is borrowed to assistant the correction of correspondence. In addition to technical contributions, we develop a realistic NCL dataset called Noise of Web (NoW), which could be a new powerful NCL benchmark where noise exists naturally. Empirical evaluations of PC$^2$ showcase marked improvements over existing state-of-the-art robust cross-modal retrieval techniques on both simulated and realistic datasets with various NCL settings. The contributed dataset and source code are released at https://github.com/alipay/PC2-NoiseofWeb.

arXiv:2407.20114공식 코드 · Papers with Code

FiCo-ITR: bridging fine-grained and coarse-grained image-text retrieval for comparative performance analysis

In the field of Image-Text Retrieval (ITR), recent advancements have leveraged large-scale Vision-Language Pretraining (VLP) for Fine-Grained (FG) instance-level retrieval, achieving high accuracy at the cost of increased computational complexity. For Coarse-Grained (CG) category-level retrieval, prominent approaches employ Cross-Modal Hashing (CMH) to prioritise efficiency, albeit at the cost of retrieval performance. Due to differences in methodologies, FG and CG models are rarely compared directly within evaluations in the literature, resulting in a lack of empirical data quantifying the retrieval performance-efficiency tradeoffs between the two. This paper addresses this gap by introducing the \texttt{FiCo-ITR} library, which standardises evaluation methodologies for both FG and CG models, facilitating direct comparisons. We conduct empirical evaluations of representative models from both subfields, analysing precision, recall, and computational complexity across varying data scales. Our findings offer new insights into the performance-efficiency trade-offs between recent representative FG and CG models, highlighting their respective strengths and limitations. These findings provide the foundation necessary to make more informed decisions regarding model selection for specific retrieval tasks and highlight avenues for future research into hybrid systems that leverage the strengths of both FG and CG approaches.

arXiv:2407.17331공식 코드 · Papers with Code

Multi-label Cluster Discrimination for Visual Representation Learning

Contrastive Language Image Pre-training (CLIP) has recently demonstrated success across various tasks due to superior feature representation empowered by image-text contrastive learning. However, the instance discrimination method used by CLIP can hardly encode the semantic structure of training data. To handle this limitation, cluster discrimination has been proposed through iterative cluster assignment and classification. Nevertheless, most cluster discrimination approaches only define a single pseudo-label for each image, neglecting multi-label signals in the image. In this paper, we propose a novel Multi-Label Cluster Discrimination method named MLCD to enhance representation learning. In the clustering step, we first cluster the large-scale LAION-400M dataset into one million centers based on off-the-shelf embedding features. Considering that natural images frequently contain multiple visual objects or attributes, we select the multiple closest centers as auxiliary class labels. In the discrimination step, we design a novel multi-label classification loss, which elegantly separates losses from positive classes and negative classes, and alleviates ambiguity on decision boundary. We validate the proposed multi-label cluster discrimination method with experiments on different scales of models and pre-training datasets. Experimental results show that our method achieves state-of-the-art performance on multiple downstream tasks including linear probe, zero-shot classification, and image-text retrieval. Code and models have been released at https://github.com/deepglint/unicom .

arXiv:2407.15239

Assessing Brittleness of Image-Text Retrieval Benchmarks from Vision-Language Models Perspective

We examine the brittleness of the image-text retrieval (ITR) evaluation pipeline with a focus on concept granularity. We start by analyzing two common benchmarks, MS-COCO and Flickr30k, and compare them with augmented, fine-grained versions, MS-COCO-FG and Flickr30k-FG, given a specified set of linguistic features capturing concept granularity. Flickr30k-FG and MS COCO-FG consistently give rise to higher scores across all the selected features. To further our understanding of the impact of granularity we consider a novel taxonomy of query perturbations. We apply these perturbations to the selected datasets. We evaluate four diverse state-of-the-art Vision-Language models on both the standard and fine-grained datasets under zero-shot conditions, with and without the applied perturbations. The results demonstrate that although perturbations generally degrade model performance, the fine-grained datasets exhibit a smaller performance drop than their standard counterparts. The relative performance drop across all setups is consistent across all models and datasets, indicating that the issue lies within the benchmarks themselves. We conclude by providing an agenda for improving ITR evaluation pipelines.

arXiv:2407.12346공식 코드 · Papers with Code

Object-Aware Query Perturbation for Cross-Modal Image-Text Retrieval

The pre-trained vision and language (V\&L) models have substantially improved the performance of cross-modal image-text retrieval. In general, however, V\&L models have limited retrieval performance for small objects because of the rough alignment between words and the small objects in the image. In contrast, it is known that human cognition is object-centric, and we pay more attention to important objects, even if they are small. To bridge this gap between the human cognition and the V\&L model's capability, we propose a cross-modal image-text retrieval framework based on ``object-aware query perturbation.'' The proposed method generates a key feature subspace of the detected objects and perturbs the corresponding queries using this subspace to improve the object awareness in the image. In our proposed method, object-aware cross-modal image-text retrieval is possible while keeping the rich expressive power and retrieval performance of existing V\&L models without additional fine-tuning. Comprehensive experiments on four public datasets show that our method outperforms conventional algorithms. Our code is publicly available at \url{https://github.com/NEC-N-SOGI/query-perturbation}.

arXiv:2407.07479

How to Make Cross Encoder a Good Teacher for Efficient Image-Text Retrieval?

Dominant dual-encoder models enable efficient image-text retrieval but suffer from limited accuracy while the cross-encoder models offer higher accuracy at the expense of efficiency. Distilling cross-modality matching knowledge from cross-encoder to dual-encoder provides a natural approach to harness their strengths. Thus we investigate the following valuable question: how to make cross-encoder a good teacher for dual-encoder? Our findings are threefold:(1) Cross-modal similarity score distribution of cross-encoder is more concentrated while the result of dual-encoder is nearly normal making vanilla logit distillation less effective. However ranking distillation remains practical as it is not affected by the score distribution.(2) Only the relative order between hard negatives conveys valid knowledge while the order information between easy negatives has little significance.(3) Maintaining the coordination between distillation loss and dual-encoder training loss is beneficial for knowledge transfer. Based on these findings we propose a novel Contrastive Partial Ranking Distillation (CPRD) method which implements the objective of mimicking relative order between hard negative samples with contrastive learning. This approach coordinates with the training of the dual-encoder effectively transferring valid knowledge from the cross-encoder to the dual-encoder. Extensive experiments on image-text retrieval and ranking tasks show that our method surpasses other distillation methods and significantly improves the accuracy of dual-encoder.

arXiv:2407.07315

CosmoCLIP: Generalizing Large Vision-Language Models for Astronomical Imaging

Existing vision-text contrastive learning models enhance representation transferability and support zero-shot prediction by matching paired image and caption embeddings while pushing unrelated pairs apart. However, astronomical image-label datasets are significantly smaller compared to general image and label datasets available from the internet. We introduce CosmoCLIP, an astronomical image-text contrastive learning framework precisely fine-tuned on the pre-trained CLIP model using SpaceNet and BLIP-based captions. SpaceNet, attained via FLARE, constitutes ~13k optimally distributed images, while BLIP acts as a rich knowledge extractor. The rich semantics derived from this SpaceNet and BLIP descriptions, when learned contrastively, enable CosmoCLIP to achieve superior generalization across various in-domain and out-of-domain tasks. Our results demonstrate that CosmoCLIP is a straightforward yet powerful framework, significantly outperforming CLIP in zero-shot classification and image-text retrieval tasks.

arXiv:2407.01081공식 코드 · Papers with Code

CVLUE: A New Benchmark Dataset for Chinese Vision-Language Understanding Evaluation

Despite the rapid development of Chinese vision-language models (VLMs), most existing Chinese vision-language (VL) datasets are constructed on Western-centric images from existing English VL datasets. The cultural bias in the images makes these datasets unsuitable for evaluating VLMs in Chinese culture. To remedy this issue, we present a new Chinese Vision- Language Understanding Evaluation (CVLUE) benchmark dataset, where the selection of object categories and images is entirely driven by Chinese native speakers, ensuring that the source images are representative of Chinese culture. The benchmark contains four distinct VL tasks ranging from image-text retrieval to visual question answering, visual grounding and visual dialogue. We present a detailed statistical analysis of CVLUE and provide a baseline performance analysis with several open-source multilingual VLMs on CVLUE and its English counterparts to reveal their performance gap between English and Chinese. Our in-depth category-level analysis reveals a lack of Chinese cultural knowledge in existing VLMs. We also find that fine-tuning on Chinese culture-related VL datasets effectively enhances VLMs' understanding of Chinese culture.

arXiv:2406.18254공식 코드 · Papers with Code

Improving the Consistency in Cross-Lingual Cross-Modal Retrieval with 1-to-K Contrastive Learning

Cross-lingual Cross-modal Retrieval (CCR) is an essential task in web search, which aims to break the barriers between modality and language simultaneously and achieves image-text retrieval in the multi-lingual scenario with a single model. In recent years, excellent progress has been made based on cross-lingual cross-modal pre-training; particularly, the methods based on contrastive learning on large-scale data have significantly improved retrieval tasks. However, these methods directly follow the existing pre-training methods in the cross-lingual or cross-modal domain, leading to two problems of inconsistency in CCR: The methods with cross-lingual style suffer from the intra-modal error propagation, resulting in inconsistent recall performance across languages in the whole dataset. The methods with cross-modal style suffer from the inter-modal optimization direction bias, resulting in inconsistent rank across languages within each instance, which cannot be reflected by Recall@K. To solve these problems, we propose a simple but effective 1-to-K contrastive learning method, which treats each language equally and eliminates error propagation and optimization bias. In addition, we propose a new evaluation metric, Mean Rank Variance (MRV), to reflect the rank inconsistency across languages within each instance. Extensive experiments on four CCR datasets show that our method improves both recall rates and MRV with smaller-scale pre-trained data, achieving the new state-of-art.

arXiv:2406.11820공식 코드 · Papers with Code

Composing Object Relations and Attributes for Image-Text Matching

We study the visual semantic embedding problem for image-text matching. Most existing work utilizes a tailored cross-attention mechanism to perform local alignment across the two image and text modalities. This is computationally expensive, even though it is more powerful than the unimodal dual-encoder approach. This work introduces a dual-encoder image-text matching model, leveraging a scene graph to represent captions with nodes for objects and attributes interconnected by relational edges. Utilizing a graph attention network, our model efficiently encodes object-attribute and object-object semantic relations, resulting in a robust and fast-performing system. Representing caption as a scene graph offers the ability to utilize the strong relational inductive bias of graph neural networks to learn object-attribute and object-object relations effectively. To train the model, we propose losses that align the image and caption both at the holistic level (image-caption) and the local level (image-object entity), which we show is key to the success of the model. Our model is termed Composition model for Object Relations and Attributes, CORA. Experimental results on two prominent image-text retrieval benchmarks, Flickr30K and MSCOCO, demonstrate that CORA outperforms existing state-of-the-art computationally expensive cross-attention methods regarding recall score while achieving fast computation speed of the dual encoder.