When Discord switched to a teen‑by‑default policy, users discovered that a simple thumb sketch with two eyes and a mouth could trick the on‑device AI age estimator into granting adult access, exposing the limits of lightweight facial analysis models.
This survey systematically reviews Embodied AI security, proposing a five‑layer taxonomy (perception, cognition, planning, action & interaction, agentic system) that organizes over 400 papers on attacks, defenses, and open challenges, and highlights overlooked vulnerabilities such as multimodal perception fusion and planning instability under jailbreak attacks.
The RaPA method introduces random parameter pruning during adversarial generation, creating diverse model variants that markedly increase the success rate of targeted transfer attacks across CNN and Transformer architectures, even against defended models and with higher computational budgets, as demonstrated on ImageNet‑compatible benchmarks.
This paper systematically evaluates 23 transferable adversarial attacks and 11 defenses on ImageNet, revealing that early methods like DI outperform many newer attacks when hyper‑parameters are fairly matched, that diffusion‑based defenses give a false sense of security, and that higher transferability often comes at the cost of reduced stealthiness.
This paper systematically evaluates 23 transferable adversarial attacks and 11 defenses on ImageNet, revealing that early DI attacks surpass newer methods when hyper‑parameters are fairly set, diffusion defenses offer false security, and higher transferability often reduces stealthiness, urging fair benchmarking and comprehensive metrics.
Researchers discovered that trivial symbols such as a colon or the word “Solution” can trigger false‑positive rewards in LLM judge models, causing GPT‑4o, Claude‑4 and LLaMA‑3‑70B to fail, and proposed a robust “Master‑RM” model that eliminates these attacks.
The 2024 IEEE‑OPPO Pan‑Terminal Security Forum, held on March 29 at Xi'an Jiaotong University, gathers leading researchers to discuss AI, IoT, blockchain, and smart contract security, presenting cutting‑edge threats, defense strategies, and future technical capabilities for intelligent terminal products.
Ant Security Lab reports four AAAI‑23 accepted papers that introduce PF‑Attack for transferable 3D adversarial point clouds, AMAP a GNN‑driven anti‑money‑laundering framework, SpikeNet a spiking‑neural‑network approach for efficient dynamic graph representation, and DP‑PSAC a per‑sample adaptive clipping method for differential privacy, each with experimental validation and expert commentary.
Meituan’s six CVPR 2022 papers advance computer vision by introducing a few‑sample model compression method, a language‑bridged video object segmentation approach, a single‑stage 3D visual grounding technique, a dynamic early‑exit image captioning system, a boosted black‑box adversarial attack, and a semi‑supervised video paragraph grounding framework.
This talk explains how to use semantic-driven captcha mechanisms to classify and manage trusted versus untrusted traffic, detailing anti‑fraud strategies, flow identification, countermeasures against simulator and protocol cracking, and proactive updates to stay ahead of black‑market attacks.
