Skip to content

Research

Page 78 of 88

EIP-7702 Phishing Attack

research note

EIP-7702 Phishing Attack

·8 min read·Minfeng Qi, Qin Wang, Ruiqiang Li et al.

The paper analyzes EIP-7702, a delegation-based Ethereum authorization scheme that allows an externally owned account (EOA) to sign a single authorization tuple which reroutes all subsequent calls …

researchethereumphishing-attackaccount-abstractiondelegation

Read note → Source paper ↗

Privacy Practices of Browser Agents

research note

Privacy Practices of Browser Agents

·7 min read·Alisha Ukani, Hamed Haddadi, Ali Shahin Shamsabadi et al.

This paper investigates the privacy practices and risks associated with browser agents—tools that automate web browsing using large language models (LLMs)

researchbrowser-agentsprivacy-evaluationweb-automationlarge-language-models

Read note → Source paper ↗

Perfect $L_p$ Sampling with Polylogarithmic Update Time

research note

Perfect $L_p$ Sampling with Polylogarithmic Update Time

·8 min read·William Swartworth, David P. Woodruff, Samson Zhou

This paper addresses the problem of perfect Lp sampling in turnstile data streams, focusing on vectors x∈{−poly(n),...,poly(n)}^n updated by turnstile operations, and sampling indices with exact pr…

researchstreaming-algorithmslp-samplingheavy-hittersturnstile-model

Read note → Source paper ↗

DLADiff: A Dual-Layer Defense Framework against Fine-Tuning and Zero-Shot Customization of Diffusion Models

research note

DLADiff: A Dual-Layer Defense Framework against Fine-Tuning and Zero-Shot Customization of Diffusion Models

·11 min read·Jun Jia, Hongyi Miao, Yingjie Zhou et al.

DLADiff addresses a gap that most prior anti-customization work leaves open: defenses for diffusion-model fine-tuning existed, but zero-shot identity customization (for example, FaceID-style method…

researchdiffusion-model-defenseanti-customizationadversarial-perturbationface-privacy

Read note → Source paper ↗

Multimodal Wireless Foundation Models

research note

Multimodal Wireless Foundation Models

·10 min read·Ahmed Aboulfotouh, Hatem Abou-Zeid

This paper tackles a concrete limitation in wireless foundation models: most prior WFMs are modality-specific, even though wireless systems expose multiple views of the same propagation event, incl…

researchmultimodal-learningwireless-foundation-modelsmasked-autoencodingself-supervised-learning

Read note → Source paper ↗

Articles are CC BY 4.0 — feel free to quote with attribution