Viral News - CybAI news

04 Jun

Disrupting Diffusion: Token-Level Attention Erasure Attack against Diffusion-based Customization

stp2y0 CommentsViral News

arXiv:2405.20584v1 Announce Type: new Abstract: With the development of diffusion-based customization methods like DreamBooth, individuals now have access to train the models that can generate their personalized images. Despite the convenience, malicious users have misused these techniques to create fake images, thereby triggering a privacy security crisis. In light of this, proactive adversarial attacks are proposed to protect users against customization. The adversarial examples are trained to distort the customization model's outputs and thus block the misuse. In this paper, we propose DisDiff (Disrupting Diffusion), a novel adversarial attack method to disrupt the diffusion model outputs. We first delve into…

04 Jun

SPOT: Text Source Prediction from Originality Score Thresholding

stp2y0 CommentsViral News

arXiv:2405.20505v1 Announce Type: new Abstract: The wide acceptance of large language models (LLMs) has unlocked new applications and social risks. Popular countermeasures aim at detecting misinformation, usually involve domain specific models trained to recognize the relevance of any information. Instead of evaluating the validity of the information, we propose to investigate LLM generated text from the perspective of trust. In this study, we define trust as the ability to know if an input text was generated by a LLM or a human. To do so, we design SPOT, an efficient method, that classifies the source of any, standalone, text input…

04 Jun

Revolutionizing Data Management: Microsoft Fabric Meets WhereScape Automation

stp2y0 CommentsViral News

Sponsored Content by WhereScape In the rapidly evolving world of data management, the integration of Microsoft Fabric with WhereScape’s automation tools marks a pivotal advancement. This synergy not only redefines the efficiencies of data operations but also empowers organizations to navigate the complexities of digital transformation with unprecedented ease. Migrating to advanced systems like Microsoft Fabric can be daunting. WhereScape data warehouse automation simplifies this transition from legacy systems, ensuring a seamless migration that minimizes downtime and maximizes data integrity. By automating the migration process, WhereScape helps businesses quickly adapt to the robust capabilities of Microsoft Fabric, facilitating a smooth…

03 Jun

Enhancing Performance for Highly Imbalanced Medical Data via Data Regularization in a Federated Learning Setting

stp2y0 CommentsViral News

[Submitted on 30 May 2024] View a PDF of the paper titled Enhancing Performance for Highly Imbalanced Medical Data via Data Regularization in a Federated Learning Setting, by Georgios Tsoumplekas and 4 other authors View PDF HTML (experimental) Abstract:The increased availability of medical data has significantly impacted healthcare by enabling the application of machine / deep learning approaches in various instances. However, medical datasets are usually small and scattered across multiple providers, suffer from high class-imbalance, and are subject to stringent data privacy constraints. In this paper, the application of a data regularization algorithm, suitable for learning under high class-imbalance,…

03 Jun

Physically Compatible 3D Object Modeling from a Single Image

stp2y0 CommentsViral News

arXiv:2405.20510v1 Announce Type: new Abstract: We present a computational framework that transforms single images into 3D physical objects. The visual geometry of a physical object in an image is determined by three orthogonal attributes: mechanical properties, external forces, and rest-shape geometry. Existing single-view 3D reconstruction methods often overlook this underlying composition, presuming rigidity or neglecting external forces. Consequently, the reconstructed objects fail to withstand real-world physical forces, resulting in instability or undesirable deformation -- diverging from their intended designs as depicted in the image. Our optimization framework addresses this by embedding physical compatibility into the reconstruction process. We explicitly decompose…

03 Jun

Transfer Q Star: Principled Decoding for LLM Alignment

stp2y0 CommentsViral News

arXiv:2405.20495v1 Announce Type: new Abstract: Aligning foundation models is essential for their safe and trustworthy deployment. However, traditional fine-tuning methods are computationally intensive and require updating billions of model parameters. A promising alternative, alignment via decoding, adjusts the response distribution directly without model updates to maximize a target reward $r$, thus providing a lightweight and adaptable framework for alignment. However, principled decoding methods rely on oracle access to an optimal Q-function ($Q^*$), which is often unavailable in practice. Hence, prior SoTA methods either approximate this $Q^*$ using $Q^{pi_{texttt{sft}}}$ (derived from the reference $texttt{SFT}$ model) or rely on short-term rewards, resulting…

03 Jun

Global Survey By McKinsey: GenAI Adoption Starting To Deliver Value

stp2y0 CommentsViral News

Investments in GenAI are beginning to create value for organizations according to a new global survey by McKinsey, a leading management consultancy firm. While 2023 was a year of investing in GenAI initiatives, 2024 is about deriving business value from this new technology. In just one and half years since OpenAI’s ChatGPT was launched, 65% of organizations are now regularly using AI, according to the McKinsey report. This is nearly double the percentage from last year’s survey. The report suggests that organizations are now using AI in more parts of the business. More than half of the respondents shared that…

03 Jun

Data + AI Strategy: Platform Focus

stp2y0 CommentsViral News

The secret to good AI is great data. As AI adoption soars, the data platform is the most important component of any enterprise's technology stack. It’s increasingly clear that Generative AI systems won’t be one monolithic, but rather a combination of many different components that must work together. And while data is one of the most important pieces, there are many other functions required for enterprises to actually deploy the models into the real-world. That’s why, when businesses are looking to build the foundational platform that will support the breadth of their data and AI needs, they should keep three core pillars…

03 Jun

Back to the Basics on Predicting Transfer Performance

stp2y0 CommentsViral News

arXiv:2405.20420v1 Announce Type: new Abstract: In the evolving landscape of deep learning, selecting the best pre-trained models from a growing number of choices is a challenge. Transferability scorers propose alleviating this scenario, but their recent proliferation, ironically, poses the challenge of their own assessment. In this work, we propose both robust benchmark guidelines for transferability scorers, and a well-founded technique to combine multiple scorers, which we show consistently improves their results. We extensively evaluate 13 scorers from literature across 11 datasets, comprising generalist, fine-grained, and medical imaging datasets. We show that few scorers match the predictive performance of the simple…

03 Jun

Slight Corruption in Pre-training Data Makes Better Diffusion Models

stp2y0 CommentsViral News

arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website. Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them. Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs. Source link lol