相关论文: RLBind: Adversarial-Invariant Cross-Modal Alignmen…

Recent unified multi-modal encoders align a wide range of modalities into a shared representation space, enabling diverse cross-modal tasks. Despite their impressive capabilities, the robustness of these models under adversarial…

We present UniBind, a flexible and efficient approach that learns a unified representation space for seven diverse modalities -- images, text, audio, point cloud, thermal, video, and event data. Existing works, eg., ImageBind, treat the…

Recent advancements in biology and chemistry have leveraged multi-modal learning, integrating molecules and their natural language descriptions to enhance drug discovery. However, current pre-training frameworks are limited to two…

Multi-modal foundation models like OpenFlamingo, LLaVA, and GPT-4 are increasingly used for various real-world tasks. Prior work has shown that these models are highly vulnerable to adversarial attacks on the vision modality. These attacks…

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…

Multimodal representation alignment is pivotal for large language models and robotics. Traditional methods are often hindered by cross-modal information discrepancies and data scarcity, leading to suboptimal alignment spaces that overlook…

Pre-trained multilingual language encoders, such as multilingual BERT and XLM-R, show great potential for zero-shot cross-lingual transfer. However, these multilingual encoders do not precisely align words and phrases across languages.…

In this paper, we propose EventBind, a novel and effective framework that unleashes the potential of vision-language models (VLMs) for event-based recognition to compensate for the lack of large-scale event-based datasets. In particular,…

Despite the success on few-shot learning problems, most meta-learned models only focus on achieving good performance on clean examples and thus easily break down when given adversarially perturbed samples. While some recent works have shown…

We present ImageBind-LLM, a multi-modality instruction tuning method of large language models (LLMs) via ImageBind. Existing works mainly focus on language and image instruction tuning, different from which, our ImageBind-LLM can respond to…

The rapid advancement of Multimodal Large Language Models (MLLMs) has extended CLIP-based frameworks to produce powerful, universal embeddings for retrieval tasks. However, existing methods primarily focus on natural images, offering…

Pretrained vision-language models (VLMs) like CLIP exhibit exceptional generalization across diverse downstream tasks. While recent studies reveal their vulnerability to adversarial attacks, research to date has primarily focused on…

Multi-modal embeddings encode texts, images, thermal images, sounds, and videos into a single embedding space, aligning representations across different modalities (e.g., associate an image of a dog with a barking sound). In this paper, we…

Medical vision-language pretraining models (VLPM) have achieved remarkable progress in fusing chest X-rays (CXR) with clinical texts, introducing image-text data binding approaches that enable zero-shot learning and downstream clinical…

Research on multi-modal learning dominantly aligns the modalities in a unified space at training, and only a single one is taken for prediction at inference. However, for a real machine, e.g., a robot, sensors could be added or removed at…

Despite recent progress in Multi-Modal Large Language Models (MLLMs), it remains challenging to integrate diverse tasks ranging from pixel-level perception to high-fidelity generation. Existing approaches often suffer from either restricted…

MLLMs have been successfully applied to multimodal embedding tasks, yet their generative reasoning capabilities remain underutilized. Directly incorporating chain-of-thought reasoning into embedding learning introduces two fundamental…

The video-language (VL) pretraining has achieved remarkable improvement in multiple downstream tasks. However, the current VL pretraining framework is hard to extend to multiple modalities (N modalities, N>=3) beyond vision and language. We…

Vision Language Models adapt well to downstream tasks but are highly vulnerable to adversarial perturbations that disrupt cross-modal semantic alignment. Existing defenses are largely unidirectional or structural, failing to exploit…

Adversarial input attacks can cause a significant shift of CLIP embeddings. This can affect the downstream robustness of models incorporating CLIP in the pipeline, such as text-to-image generative models or large vision language models.…