Tootfinder

Vision-Language-Vision Auto-Encoder: Scalable Knowledge Distillation from Diffusion Models
Tiezheng Zhang, Yitong Li, Yu-cheng Chou, Jieneng Chen, Alan Yuille, Chen Wei, Junfei Xiao
https://arxiv.org/abs/2507.07104

Vision-Language-Vision Auto-Encoder: Scalable Knowledge Distillation from Diffusion Models
Building state-of-the-art Vision-Language Models (VLMs) with strong captioning capabilities typically necessitates training on billions of high-quality image-text pairs, requiring millions of GPU hours. This paper introduces the Vision-Language-Vision (VLV) auto-encoder framework, which strategically leverages key pretrained components: a vision encoder, the decoder of a Text-to-Image (T2I) diffusion model, and subsequently, a Large Language Model (LLM). Specifically, we establish an informatio…

Ein großer Meilenstein für die Versorgung mit Medikamenten wurde gerade in Frankreich gegangen. Die Pharma-Firmen, die die Produktion von essentiellen Medikamenten einstellen wollen, müssen eine Alternativ-Versorgung sicherstellen oder alle Rechte an eine öffentliche Einrichtung abgeben. Das sollte in Deutschland unbedingt Nachahmung finden!

Le gouvernement français impose des obligations strictes aux laboratoires qui veulent cesser de produire certa
Pour éviter les pénuries de médicaments d'intérêt thérapeutique majeur, un décret impose désormais des obligations très strictes aux laboratoires qui voudraient arrêter de les produire.

JPS: Jailbreak Multimodal Large Language Models with Collaborative Visual Perturbation and Textual Steering
Renmiao Chen, Shiyao Cui, Xuancheng Huang, Chengwei Pan, Victor Shea-Jay Huang, QingLin Zhang, Xuan Ouyang, Zhexin Zhang, Hongning Wang, Minlie Huang
https://arxiv.org/abs/2508.05087

JPS: Jailbreak Multimodal Large Language Models with Collaborative Visual Perturbation and Textual Steering
Jailbreak attacks against multimodal large language Models (MLLMs) are a significant research focus. Current research predominantly focuses on maximizing attack success rate (ASR), often overlooking whether the generated responses actually fulfill the attacker's malicious intent. This oversight frequently leads to low-quality outputs that bypass safety filters but lack substantial harmful content. To address this gap, we propose JPS, \underline{J}ailbreak MLLMs with collaborative visual \underl…

From Contrast to Commonality: Audio Commonality Captioning for Enhanced Audio-Text Cross-modal Understanding in Multimodal LLMs
Yuhang Jia, Xu Zhang, Yong Qin
https://arxiv.org/abs/2508.01659

From Contrast to Commonality: Audio Commonality Captioning for Enhanced Audio-Text Cross-modal Understanding in Multimodal LLMs
Audio Captioning (AC) plays a pivotal role in enhancing audio-text cross-modal understanding during the pretraining and finetuning of multimodal large language models (MLLMs). To further strengthen this alignment, recent works have proposed Audio Difference Captioning (ADC), which takes multiple audio inputs and encourages the model to describe their differences, thereby promoting fine-grained audio discrimination. However, despite its effectiveness in enabling difference-telling and detailed d…

A Note on Chaos in Hayward Black Holes with String Fluids
Aditya Singh, Ashes Modak, Binata Panda
https://arxiv.org/abs/2507.02716 https://

A Note on Chaos in Hayward Black Holes with String Fluids
In this work, we examine the onset of thermodynamic chaos in Hayward AdS black holes with string fluids, emphasizing the effects of temporal and spatially periodic perturbations. We apply Melnikov's approach to examine the perturbed Hamiltonian dynamics and detect the onset of chaotic behavior within the spinodal regime of the $(P-v)$ plot. In the case of temporal perturbations induced by thermal quenches, chaos occurs for perturbation amplitude $γ$ exceeding a critical threshold that can be d…

On the Feasibility of Poisoning Text-to-Image AI Models via Adversarial Mislabeling
Stanley Wu, Ronik Bhaskar, Anna Yoo Jeong Ha, Shawn Shan, Haitao Zheng, Ben Y. Zhao
https://arxiv.org/abs/2506.21874

On the Feasibility of Poisoning Text-to-Image AI Models via Adversarial Mislabeling
Today's text-to-image generative models are trained on millions of images sourced from the Internet, each paired with a detailed caption produced by Vision-Language Models (VLMs). This part of the training pipeline is critical for supplying the models with large volumes of high-quality image-caption pairs during training. However, recent work suggests that VLMs are vulnerable to stealthy adversarial attacks, where adversarial perturbations are added to images to mislead the VLMs into producing …

Can Large Pretrained Depth Estimation Models Help With Image Dehazing?
Hongfei Zhang, Kun Zhou, Ruizheng Wu, Jiangbo Lu
https://arxiv.org/abs/2508.00698 https://

Can Large Pretrained Depth Estimation Models Help With Image Dehazing?
Image dehazing remains a challenging problem due to the spatially varying nature of haze in real-world scenes. While existing methods have demonstrated the promise of large-scale pretrained models for image dehazing, their architecture-specific designs hinder adaptability across diverse scenarios with different accuracy and efficiency requirements. In this work, we systematically investigate the generalization capability of pretrained depth representations-learned from millions of diverse image…

Applying Vision Transformers on Spectral Analysis of Astronomical Objects
Luis Felipe Strano Moraes, Ignacio Becker, Pavlos Protopapas, Guillermo Cabrera-Vives
https://arxiv.org/abs/2506.00294

Applying Vision Transformers on Spectral Analysis of Astronomical Objects
We apply pre-trained Vision Transformers (ViTs), originally developed for image recognition, to the analysis of astronomical spectral data. By converting traditional one-dimensional spectra into two-dimensional image representations, we enable ViTs to capture both local and global spectral features through spatial self-attention. We fine-tune a ViT pretrained on ImageNet using millions of spectra from the SDSS and LAMOST surveys, represented as spectral plots. Our model is evaluated on key task…

Existence and decay for a Grushin problem in $\mathbb{R}^N$ with singular, convective, critical reaction
Laura Baldelli, Paolo Malanchini, Simone Secchi
https://arxiv.org/abs/2506.22177

Existence and decay for a Grushin problem in $\mathbb{R}^N$ with singular, convective, critical reaction
We establish an existence result for a problem set in the whole Euclidean space involving the Grushin operator and featuring a critical term perturbed by a singular, convective reaction. Our approach combines variational methods, truncation techniques, and concentration-compactness arguments, together with set-valued analysis and fixed point theory. Additionally, we prove the decay at infinity of solutions in the absence of the convective term. The result is new even in the case where more than…

MetaCLIP 2: A Worldwide Scaling Recipe
Yung-Sung Chuang, Yang Li, Dong Wang, Ching-Feng Yeh, Kehan Lyu, Ramya Raghavendra, James Glass, Lifei Huang, Jason Weston, Luke Zettlemoyer, Xinlei Chen, Zhuang Liu, Saining Xie, Wen-tau Yih, Shang-Wen Li, Hu Xu
https://arxiv.org/abs/2507.22062

MetaCLIP 2: A Worldwide Scaling Recipe
Contrastive Language-Image Pretraining (CLIP) is a popular foundation model, supporting from zero-shot classification, retrieval to encoders for multimodal large language models (MLLMs). Although CLIP is successfully trained on billion-scale image-text pairs from the English world, scaling CLIP's training further to learning from the worldwide web data is still challenging: (1) no curation method is available to handle data points from non-English world; (2) the English performance from existin…