Tootfinder

Network Structures as an Attack Surface: Topology-Based Privacy Leakage in Federated Learning
Murtaza Rangwala, Richard O. Sinnott, Rajkumar Buyya
https://arxiv.org/abs/2506.19260

Network Structures as an Attack Surface: Topology-Based Privacy Leakage in Federated Learning
Federated learning systems increasingly rely on diverse network topologies to address scalability and organizational constraints. While existing privacy research focuses on gradient-based attacks, the privacy implications of network topology knowledge remain critically understudied. We conduct the first comprehensive analysis of topology-based privacy leakage across realistic adversarial knowledge scenarios, demonstrating that adversaries with varying degrees of structural knowledge can infer s…

Oye @… hazte autopromo del articulo este de "Renfe se atrinchera en Europa" para que podamos retwiteartelo no? Jejeje

The Irish Data Protection Commission says Meta told it that WhatsApp ads "won't be launching" in the EU "until 2026"; WhatsApp says ads are coming "gradually" (Ellen O'Regan/Politico)
https://www.politico.eu/article/whatsapp-m

WhatsApp won’t roll out ads in EU until 2026
Meta’s advertising plan has triggered privacy concerns in Europe.

Oh WhatsApp now brings Ads to the stories... Too bad that still so many people in my social environment use it
https://www.theverge.com/news/687519/whatsapp-launch-advertising-status-updates

WhatsApp is officially getting ads
WhatsApp is getting ads, as Meta has announced that it will now allow businesses to promote their products and services through its status feature.

Experimental Design for Semiparametric Bandits
Seok-Jin Kim, Gi-Soo Kim, Min-hwan Oh
https://arxiv.org/abs/2506.13390 https://arxiv.o…

Experimental Design for Semiparametric Bandits
We study finite-armed semiparametric bandits, where each arm's reward combines a linear component with an unknown, potentially adversarial shift. This model strictly generalizes classical linear bandits and reflects complexities common in practice. We propose the first experimental-design approach that simultaneously offers a sharp regret bound, a PAC bound, and a best-arm identification guarantee. Our method attains the minimax regret $\tilde{O}(\sqrt{dT})$, matching the known lower bound for …

PuDHammer: Experimental Analysis of Read Disturbance Effects of Processing-using-DRAM in Real DRAM Chips
Ismail Emir Yuksel, Akash Sood, Ataberk Olgun, O\u{g}uzhan Canpolat, Haocong Luo, F. Nisa Bostanc{\i}, Mohammad Sadrosadati, A. Giray Ya\u{g}l{\i}k\c{c}{\i}, Onur Mutlu
https://arxiv.org/abs/2506.12947

PuDHammer: Experimental Analysis of Read Disturbance Effects of Processing-using-DRAM in Real DRAM Chips
Processing-using-DRAM (PuD) is a promising paradigm for alleviating the data movement bottleneck using DRAM's massive internal parallelism and bandwidth to execute very wide operations. Performing a PuD operation involves activating multiple DRAM rows in quick succession or simultaneously, i.e., multiple-row activation. Multiple-row activation is fundamentally different from conventional memory access patterns that activate one DRAM row at a time. However, repeatedly activating even one DRAM ro…

One Patch to Rule Them All: Transforming Static Patches into Dynamic Attacks in the Physical World
Xingshuo Han, Chen Ling, Shiyi Yao, Haozhao Wang, Hangcheng Liu, Yutong Wu, Shengmin Xu, Changhai Ou, Xinyi Huang, Tianwei Zhang
https://arxiv.org/abs/2506.08482

One Patch to Rule Them All: Transforming Static Patches into Dynamic Attacks in the Physical World
Numerous methods have been proposed to generate physical adversarial patches (PAPs) against real-world machine learning systems. However, each existing PAP typically supports only a single, fixed attack goal, and switching to a different objective requires re-generating and re-deploying a new PAP. This rigidity limits their practicality in dynamic environments like autonomous driving, where traffic conditions and attack goals can change rapidly. For example, if no obstacles are present around t…

EasyDRAM: An FPGA-based Infrastructure for Fast and Accurate End-to-End Evaluation of Emerging DRAM Techniques
O\u{g}uzhan Canpolat, Ataberk Olgun, David Novo, O\u{g}uz Ergin, Onur Mutlu
https://arxiv.org/abs/2506.10441

EasyDRAM: An FPGA-based Infrastructure for Fast and Accurate End-to-End Evaluation of Emerging DRAM Techniques
DRAM is a critical component of modern computing systems. Recent works propose numerous techniques (that we call DRAM techniques) to enhance DRAM-based computing systems' throughput, reliability, and computing capabilities (e.g., in-DRAM bulk data copy). Evaluating the system-wide benefits of DRAM techniques is challenging as they often require modifications across multiple layers of the computing stack. Prior works propose FPGA-based platforms for rapid end-to-end evaluation of DRAM techniques…

This https://arxiv.org/abs/2505.18889 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_csCR_…

Security Concerns for Large Language Models: A Survey
Large Language Models (LLMs) such as GPT-4 (and its recent iterations like GPT-4o and the GPT-4.1 series), Google's Gemini, Anthropic's Claude 3 models, and xAI's Grok have caused a revolution in natural language processing, but their capabilities also introduce new security vulnerabilities. In this survey, we provide a comprehensive overview of the emerging security concerns around LLMs, categorizing threats into prompt injection and jailbreaking, adversarial attacks (including input perturbat…

This https://arxiv.org/abs/2501.18626 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_csCR_…

The TIP of the Iceberg: Revealing a Hidden Class of Task-in-Prompt Adversarial Attacks on LLMs
We present a novel class of jailbreak adversarial attacks on LLMs, termed Task-in-Prompt (TIP) attacks. Our approach embeds sequence-to-sequence tasks (e.g., cipher decoding, riddles, code execution) into the model's prompt to indirectly generate prohibited inputs. To systematically assess the effectiveness of these attacks, we introduce the PHRYGE benchmark. We demonstrate that our techniques successfully circumvent safeguards in six state-of-the-art language models, including GPT-4o and LLaMA…