PyTorchKR Papers

최신 논문

총 1,759편
태그·기간 등 상세 필터

카테고리 탐색

로그인 후 이 태그 구독

arXiv:2502.01576공식 코드 · Papers with Code

Robust-LLaVA: On the Effectiveness of Large-Scale Robust Image Encoders for Multi-modal Large Language Models

Multi-modal Large Language Models (MLLMs) excel in vision-language tasks but remain vulnerable to visual adversarial perturbations that can induce hallucinations, manipulate responses, or bypass safety mechanisms. Existing methods seek to mitigate these risks by applying constrained adversarial fine-tuning to CLIP vision encoders on ImageNet-scale data, ensuring their generalization ability is preserved. However, this limited adversarial training restricts robustness and broader generalization. In this work, we explore an alternative approach of leveraging existing vision classification models that have been adversarially pre-trained on large-scale data. Our analysis reveals two principal contributions: (1) the extensive scale and diversity of adversarial pre-training enables these models to demonstrate superior robustness against diverse adversarial threats, ranging from imperceptible perturbations to advanced jailbreaking attempts, without requiring additional adversarial training, and (2) end-to-end MLLM integration with these robust models facilitates enhanced adaptation of language components to robust visual features, outperforming existing plug-and-play methodologies on complex reasoning tasks. Through systematic evaluation across visual question-answering, image captioning, and jail-break attacks, we demonstrate that MLLMs trained with these robust models achieve superior adversarial robustness while maintaining favorable clean performance. Our framework achieves 2x and 1.5x average robustness gains in captioning and VQA tasks, respectively, and delivers over 10% improvement against jailbreak attacks. Code and pretrained models will be available at https://github.com/HashmatShadab/Robust-LLaVA.

arXiv:2502.02589

COCONut-PanCap: Joint Panoptic Segmentation and Grounded Captions for Fine-Grained Understanding and Generation

This paper introduces the COCONut-PanCap dataset, created to enhance panoptic segmentation and grounded image captioning. Building upon the COCO dataset with advanced COCONut panoptic masks, this dataset aims to overcome limitations in existing image-text datasets that often lack detailed, scene-comprehensive descriptions. The COCONut-PanCap dataset incorporates fine-grained, region-level captions grounded in panoptic segmentation masks, ensuring consistency and improving the detail of generated captions. Through human-edited, densely annotated descriptions, COCONut-PanCap supports improved training of vision-language models (VLMs) for image understanding and generative models for text-to-image tasks. Experimental results demonstrate that COCONut-PanCap significantly boosts performance across understanding and generation tasks, offering complementary benefits to large-scale datasets. This dataset sets a new benchmark for evaluating models on joint panoptic segmentation and grounded captioning tasks, addressing the need for high-quality, detailed image-text annotations in multi-modal learning.

arXiv:2501.18362

MedXpertQA: Benchmarking Expert-Level Medical Reasoning and Understanding

We introduce MedXpertQA, a highly challenging and comprehensive benchmark to evaluate expert-level medical knowledge and advanced reasoning. MedXpertQA includes 4,460 questions spanning 17 specialties and 11 body systems. It includes two subsets, Text for text evaluation and MM for multimodal evaluation. Notably, MM introduces expert-level exam questions with diverse images and rich clinical information, including patient records and examination results, setting it apart from traditional medical multimodal benchmarks with simple QA pairs generated from image captions. MedXpertQA applies rigorous filtering and augmentation to address the insufficient difficulty of existing benchmarks like MedQA, and incorporates specialty board questions to improve clinical relevance and comprehensiveness. We perform data synthesis to mitigate data leakage risk and conduct multiple rounds of expert reviews to ensure accuracy and reliability. We evaluate 16 leading models on MedXpertQA. Moreover, medicine is deeply connected to real-world decision-making, providing a rich and representative setting for assessing reasoning abilities beyond mathematics and code. To this end, we develop a reasoning-oriented subset to facilitate the assessment of o1-like models.

arXiv:2501.14828

An Ensemble Model with Attention Based Mechanism for Image Captioning

Image captioning creates informative text from an input image by creating a relationship between the words and the actual content of an image. Recently, deep learning models that utilize transformers have been the most successful in automatically generating image captions. The capabilities of transformer networks have led to notable progress in several activities related to vision. In this paper, we thoroughly examine transformer models, emphasizing the critical role that attention mechanisms play. The proposed model uses a transformer encoder-decoder architecture to create textual captions and a deep learning convolutional neural network to extract features from the images. To create the captions, we present a novel ensemble learning framework that improves the richness of the generated captions by utilizing several deep neural network architectures based on a voting mechanism that chooses the caption with the highest bilingual evaluation understudy (BLEU) score. The proposed model was evaluated using publicly available datasets. Using the Flickr8K dataset, the proposed model achieved the highest BLEU-[1-3] scores with rates of 0.728, 0.495, and 0.323, respectively. The suggested model outperformed the latest methods in Flickr30k datasets, determined by BLEU-[1-4] scores with rates of 0.798, 0.561, 0.387, and 0.269, respectively. The model efficacy was also obtained by the Semantic propositional image caption evaluation (SPICE) metric with a scoring rate of 0.164 for the Flicker8k dataset and 0.387 for the Flicker30k. Finally, ensemble learning significantly advances the process of image captioning and, hence, can be leveraged in various applications across different domains.

arXiv:2501.13851공식 코드 · Papers with Code

Large Vision-Language Models for Knowledge-Grounded Data Annotation of Memes

Memes have emerged as a powerful form of communication, integrating visual and textual elements to convey humor, satire, and cultural messages. Existing research has focused primarily on aspects such as emotion classification, meme generation, propagation, interpretation, figurative language, and sociolinguistics, but has often overlooked deeper meme comprehension and meme-text retrieval. To address these gaps, this study introduces ClassicMemes-50-templates (CM50), a large-scale dataset consisting of over 33,000 memes, centered around 50 popular meme templates. We also present an automated knowledge-grounded annotation pipeline leveraging large vision-language models to produce high-quality image captions, meme captions, and literary device labels overcoming the labor intensive demands of manual annotation. Additionally, we propose a meme-text retrieval CLIP model (mtrCLIP) that utilizes cross-modal embedding to enhance meme analysis, significantly improving retrieval performance. Our contributions include:(1) a novel dataset for large-scale meme study, (2) a scalable meme annotation framework, and (3) a fine-tuned CLIP for meme-text retrieval, all aimed at advancing the understanding and analysis of memes at scale.

arXiv:2501.12206공식 코드 · Papers with Code

PAINT: Paying Attention to INformed Tokens to Mitigate Hallucination in Large Vision-Language Model

Large Vision Language Models (LVLMs) have demonstrated remarkable capabilities in understanding and describing visual content, achieving state-of-the-art performance across various vision-language tasks. However, these models often generate descriptions containing objects or details that are absent in the input image, a phenomenon commonly known as hallucination. Our work investigates the key reasons behind this issue by analyzing the pattern of self-attention in transformer layers. We find that hallucinations often arise from the progressive weakening of attention weight to visual tokens in the deeper layers of the LLM. Some previous works naively boost the attention of all visual tokens to mitigate this issue, resulting in suboptimal hallucination reduction. To address this, we identify two critical sets of visual tokens that facilitate the transfer of visual information from the vision encoder to the LLM. Local tokens encode grounded information about objects present in an image, while summary tokens capture the overall aggregated representation of the image. Importantly, these two sets of tokens require different levels of weight enhancement. To this end, we propose \textbf{PAINT} (\textbf{P}aying \textbf{A}ttention to \textbf{IN}formed \textbf{T}okens), a plug-and-play framework that intervenes in the self-attention mechanism of the LLM, selectively boosting the attention weights of local and summary tokens with experimentally learned margins. Evaluation on the MSCOCO image captioning dataset demonstrate that our approach reduces hallucination rates by up to 62.3\% compared to baseline models while maintaining accuracy. Code is available at \href{https://github.com/hasanar1f/PAINT}{https://github.com/hasanar1f/PAINT}

arXiv:2412.00095

OPCap:Object-aware Prompting Captioning

In the field of image captioning, the phenomenon where missing or nonexistent objects are used to explain an image is referred to as object bias (or hallucination). To mitigate this issue, we propose a target-aware prompting strategy. This method first extracts object labels and their spatial information from the image using an object detector. Then, an attribute predictor further refines the semantic features of the objects. These refined features are subsequently integrated and fed into the decoder, enhancing the model's understanding of the image context. Experimental results on the COCO and nocaps datasets demonstrate that OPCap effectively mitigates hallucination and significantly improves the quality of generated captions.

arXiv:2501.09446

Double Visual Defense: Adversarial Pre-training and Instruction Tuning for Improving Vision-Language Model Robustness

This paper investigates the robustness of vision-language models against adversarial visual perturbations and introduces a novel ``double visual defense" to enhance this robustness. Unlike previous approaches that resort to lightweight adversarial fine-tuning of a pre-trained CLIP model, we perform large-scale adversarial vision-language pre-training from scratch using web-scale data. We then strengthen the defense by incorporating adversarial visual instruction tuning. The resulting models from each stage, $\Delta$CLIP and $\Delta^2$LLaVA, show substantially enhanced zero-shot robustness and set a new state-of-the-art in adversarial defense for vision-language models. For example, the adversarial robustness of $\Delta$CLIP surpasses that of the previous best models on ImageNet-1k by ~20%. %For example, $\Delta$CLIP surpasses the previous best models on ImageNet-1k by ~20% in terms of adversarial robustness. Similarly, compared to prior art, $\Delta^2$LLaVA brings a ~30% robustness improvement to image captioning task and a ~20% robustness improvement to visual question answering task. Furthermore, our models exhibit stronger zero-shot recognition capability, fewer hallucinations, and superior reasoning performance compared to baselines. Our project page is https://doublevisualdefense.github.io/.

arXiv:2501.09155

VCRScore: Image captioning metric based on V\&L Transformers, CLIP, and precision-recall

Image captioning has become an essential Vision & Language research task. It is about predicting the most accurate caption given a specific image or video. The research community has achieved impressive results by continuously proposing new models and approaches to improve the overall model's performance. Nevertheless, despite increasing proposals, the performance metrics used to measure their advances have remained practically untouched through the years. A probe of that, nowadays metrics like BLEU, METEOR, CIDEr, and ROUGE are still very used, aside from more sophisticated metrics such as BertScore and ClipScore. Hence, it is essential to adjust how are measure the advances, limitations, and scopes of the new image captioning proposals, as well as to adapt new metrics to these new advanced image captioning approaches. This work proposes a new evaluation metric for the image captioning problem. To do that, first, it was generated a human-labeled dataset to assess to which degree the captions correlate with the image's content. Taking these human scores as ground truth, we propose a new metric, and compare it with several well-known metrics, from classical to newer ones. Outperformed results were also found, and interesting insights were presented and discussed.

arXiv:2501.09555공식 코드 · Papers with Code

Text-driven Adaptation of Foundation Models for Few-shot Surgical Workflow Analysis

Purpose: Surgical workflow analysis is crucial for improving surgical efficiency and safety. However, previous studies rely heavily on large-scale annotated datasets, posing challenges in cost, scalability, and reliance on expert annotations. To address this, we propose Surg-FTDA (Few-shot Text-driven Adaptation), designed to handle various surgical workflow analysis tasks with minimal paired image-label data. Methods: Our approach has two key components. First, Few-shot selection-based modality alignment selects a small subset of images and aligns their embeddings with text embeddings from the downstream task, bridging the modality gap. Second, Text-driven adaptation leverages only text data to train a decoder, eliminating the need for paired image-text data. This decoder is then applied to aligned image embeddings, enabling image-related tasks without explicit image-text pairs. Results: We evaluate our approach to generative tasks (image captioning) and discriminative tasks (triplet recognition and phase recognition). Results show that Surg-FTDA outperforms baselines and generalizes well across downstream tasks. Conclusion: We propose a text-driven adaptation approach that mitigates the modality gap and handles multiple downstream tasks in surgical workflow analysis, with minimal reliance on large annotated datasets. The code and dataset will be released in https://github.com/CAMMA-public/Surg-FTDA

arXiv:2501.09291공식 코드 · Papers with Code

LAVCap: LLM-based Audio-Visual Captioning using Optimal Transport

Automated audio captioning is a task that generates textual descriptions for audio content, and recent studies have explored using visual information to enhance captioning quality. However, current methods often fail to effectively fuse audio and visual data, missing important semantic cues from each modality. To address this, we introduce LAVCap, a large language model (LLM)-based audio-visual captioning framework that effectively integrates visual information with audio to improve audio captioning performance. LAVCap employs an optimal transport-based alignment loss to bridge the modality gap between audio and visual features, enabling more effective semantic extraction. Additionally, we propose an optimal transport attention module that enhances audio-visual fusion using an optimal transport assignment map. Combined with the optimal training strategy, experimental results demonstrate that each component of our framework is effective. LAVCap outperforms existing state-of-the-art methods on the AudioCaps dataset, without relying on large datasets or post-processing. Code is available at https://github.com/NAVER-INTEL-Co-Lab/gaudi-lavcap.

arXiv:2501.06828

GeoPix: Multi-Modal Large Language Model for Pixel-level Image Understanding in Remote Sensing

Multi-modal large language models (MLLMs) have achieved remarkable success in image- and region-level remote sensing (RS) image understanding tasks, such as image captioning, visual question answering, and visual grounding. However, existing RS MLLMs lack the pixel-level dialogue capability, which involves responding to user instructions with segmentation masks for specific instances. In this paper, we propose GeoPix, a RS MLLM that extends image understanding capabilities to the pixel level. This is achieved by equipping the MLLM with a mask predictor, which transforms visual features from the vision encoder into masks conditioned on the LLM's segmentation token embeddings. To facilitate the segmentation of multi-scale objects in RS imagery, a class-wise learnable memory module is integrated into the mask predictor to capture and store class-wise geo-context at the instance level across the entire dataset. In addition, to address the absence of large-scale datasets for training pixel-level RS MLLMs, we construct the GeoPixInstruct dataset, comprising 65,463 images and 140,412 instances, with each instance annotated with text descriptions, bounding boxes, and masks. Furthermore, we develop a two-stage training strategy to balance the distinct requirements of text generation and masks prediction in multi-modal multi-task optimization. Extensive experiments verify the effectiveness and superiority of GeoPix in pixel-level segmentation tasks, while also maintaining competitive performance in image- and region-level benchmarks.

arXiv:2501.07525공식 코드 · Papers with Code

RadAlign: Advancing Radiology Report Generation with Vision-Language Concept Alignment

Automated chest radiographs interpretation requires both accurate disease classification and detailed radiology report generation, presenting a significant challenge in the clinical workflow. Current approaches either focus on classification accuracy at the expense of interpretability or generate detailed but potentially unreliable reports through image captioning techniques. In this study, we present RadAlign, a novel framework that combines the predictive accuracy of vision-language models (VLMs) with the reasoning capabilities of large language models (LLMs). Inspired by the radiologist's workflow, RadAlign first employs a specialized VLM to align visual features with key medical concepts, achieving superior disease classification with an average AUC of 0.885 across multiple diseases. These recognized medical conditions, represented as text-based concepts in the aligned visual-language space, are then used to prompt LLM-based report generation. Enhanced by a retrieval-augmented generation mechanism that grounds outputs in similar historical cases, RadAlign delivers superior report quality with a GREEN score of 0.678, outperforming state-of-the-art methods' 0.634. Our framework maintains strong clinical interpretability while reducing hallucinations, advancing automated medical imaging and report analysis through integrated predictive and generative AI. Code is available at https://github.com/difeigu/RadAlign.

arXiv:2501.05901공식 코드 · Papers with Code

Valley2: Exploring Multimodal Models with Scalable Vision-Language Design

Recently, vision-language models have made remarkable progress, demonstrating outstanding capabilities in various tasks such as image captioning and video understanding. We introduce Valley2, a novel multimodal large language model designed to enhance performance across all domains and extend the boundaries of practical applications in e-commerce and short video scenarios. Notably, Valley2 achieves state-of-the-art (SOTA) performance on e-commerce benchmarks, surpassing open-source models of similar size by a large margin (79.66 vs. 72.76). Additionally, Valley2 ranks second on the OpenCompass leaderboard among models with fewer than 10B parameters, with an impressive average score of 67.4. The code and model weights are open-sourced at https://github.com/bytedance/Valley.

arXiv:2501.04513

Improving Image Captioning by Mimicking Human Reformulation Feedback at Inference-time

Incorporating automatically predicted human feedback into the process of training generative models has attracted substantial recent interest, while feedback at inference time has received less attention. The typical feedback at training time, i.e., preferences of choice given two samples, does not naturally transfer to the inference phase. We introduce a novel type of feedback -- caption reformulations -- and train models to mimic reformulation feedback based on human annotations. Our method does not require training the image captioning model itself, thereby demanding substantially less computational effort. We experiment with two types of reformulation feedback: first, we collect a dataset of human reformulations that correct errors in the generated captions. We find that incorporating reformulation models trained on this data into the inference phase of existing image captioning models results in improved captions, especially when the original captions are of low quality. We apply our method to non-English image captioning, a domain where robust models are less prevalent, and gain substantial improvement. Second, we apply reformulations to style transfer. Quantitative evaluations reveal state-of-the-art performance on German image captioning and English style transfer, while human validation with a detailed comparative framework exposes the specific axes of improvement.

arXiv:2501.03567

Evaluating Image Caption via Cycle-consistent Text-to-Image Generation

Evaluating image captions typically relies on reference captions, which are costly to obtain and exhibit significant diversity and subjectivity. While reference-free evaluation metrics have been proposed, most focus on cross-modal evaluation between captions and images. Recent research has revealed that the modality gap generally exists in the representation of contrastive learning-based multi-modal systems, undermining the reliability of cross-modality metrics like CLIPScore. In this paper, we propose CAMScore, a cyclic reference-free automatic evaluation metric for image captioning models. To circumvent the aforementioned modality gap, CAMScore utilizes a text-to-image model to generate images from captions and subsequently evaluates these generated images against the original images. Furthermore, to provide fine-grained information for a more comprehensive evaluation, we design a three-level evaluation framework for CAMScore that encompasses pixel-level, semantic-level, and objective-level perspectives. Extensive experiment results across multiple benchmark datasets show that CAMScore achieves a superior correlation with human judgments compared to existing reference-based and reference-free metrics, demonstrating the effectiveness of the framework.

arXiv:2501.02669공식 코드 · Papers with Code

Generalizing from SIMPLE to HARD Visual Reasoning: Can We Mitigate Modality Imbalance in VLMs?

While Vision Language Models (VLMs) are impressive in tasks such as visual question answering (VQA) and image captioning, their ability to apply multi-step reasoning to images has lagged, giving rise to perceptions of modality imbalance or brittleness. Towards systematic study of such issues, we introduce a synthetic framework for assessing the ability of VLMs to perform algorithmic visual reasoning (AVR), comprising three tasks: Table Readout, Grid Navigation, and Visual Analogy. Each has two levels of difficulty, SIMPLE and HARD, and even the SIMPLE versions are difficult for frontier VLMs. We seek strategies for training on the SIMPLE version of the tasks that improve performance on the corresponding HARD task, i.e., S2H generalization. This synthetic framework, where each task also has a text-only version, allows a quantification of the modality imbalance, and how it is impacted by training strategy. Ablations highlight the importance of explicit image-to-text conversion in promoting S2H generalization when using auto-regressive training. We also report results of mechanistic study of this phenomenon, including a measure of gradient alignment that seems to identify training strategies that promote better S2H generalization.

arXiv:2501.02570공식 코드 · Papers with Code

Decoding fMRI Data into Captions using Prefix Language Modeling

With the advancements in Large Language and Latent Diffusion models, brain decoding has achieved remarkable results in recent years. The works on the NSD dataset, with stimuli images from the COCO dataset, leverage the embeddings from the CLIP model for image reconstruction and GIT for captioning. However, the current captioning approach introduces the challenge of potential data contamination given that the GIT model was trained on the COCO dataset. In this work, we present an alternative method for decoding brain signals into image captions by predicting a DINOv2 model's embedding of an image from the corresponding fMRI signal and then providing its [CLS] token as the prefix to the GPT-2 language model which decreases computational requirements considerably. Additionally, instead of commonly used Linear Regression, we explore 3D Convolutional Neural Network mapping of fMRI signals to image embedding space for better accounting positional information of voxels.

arXiv:2501.01834

MoColl: Agent-Based Specific and General Model Collaboration for Image Captioning

Image captioning is a critical task at the intersection of computer vision and natural language processing, with wide-ranging applications across various domains. For complex tasks such as diagnostic report generation, deep learning models require not only domain-specific image-caption datasets but also the incorporation of relevant general knowledge to provide contextual accuracy. Existing approaches exhibit inherent limitations: specialized models excel in capturing domain-specific details but lack generalization, while vision-language models (VLMs) built on large language models (LLMs) leverage general knowledge but struggle with domain-specific adaptation. To address these limitations, this paper proposes a novel agent-enhanced model collaboration framework, which we call MoColl, designed to effectively integrate domain-specific and general knowledge. Specifically, our approach is to decompose complex image captioning tasks into a series of interconnected question-answer subtasks. A trainable visual question answering (VQA) model is employed as a specialized tool to focus on domain-specific visual analysis, answering task-specific questions based on image content. Concurrently, an LLM-based agent with general knowledge formulates these questions and synthesizes the resulting question-answer pairs into coherent captions. Beyond its role in leveraging the VQA model, the agent further guides its training to enhance its domain-specific capabilities. Experimental results on radiology report generation validate the effectiveness of the proposed framework, demonstrating significant improvements in the quality of generated reports.

arXiv:2501.00437

Unleashing Text-to-Image Diffusion Prior for Zero-Shot Image Captioning

Recently, zero-shot image captioning has gained increasing attention, where only text data is available for training. The remarkable progress in text-to-image diffusion model presents the potential to resolve this task by employing synthetic image-caption pairs generated by this pre-trained prior. Nonetheless, the defective details in the salient regions of the synthetic images introduce semantic misalignment between the synthetic image and text, leading to compromised results. To address this challenge, we propose a novel Patch-wise Cross-modal feature Mix-up (PCM) mechanism to adaptively mitigate the unfaithful contents in a fine-grained manner during training, which can be integrated into most of encoder-decoder frameworks, introducing our PCM-Net. Specifically, for each input image, salient visual concepts in the image are first detected considering the image-text similarity in CLIP space. Next, the patch-wise visual features of the input image are selectively fused with the textual features of the salient visual concepts, leading to a mixed-up feature map with less defective content. Finally, a visual-semantic encoder is exploited to refine the derived feature map, which is further incorporated into the sentence decoder for caption generation. Additionally, to facilitate the model training with synthetic data, a novel CLIP-weighted cross-entropy loss is devised to prioritize the high-quality image-text pairs over the low-quality counterparts. Extensive experiments on MSCOCO and Flickr30k datasets demonstrate the superiority of our PCM-Net compared with state-of-the-art VLMs-based approaches. It is noteworthy that our PCM-Net ranks first in both in-domain and cross-domain zero-shot image captioning. The synthetic dataset SynthImgCap and code are available at https://jianjieluo.github.io/SynthImgCap.