Tagged articles
adversarial attacks
17 articles · Page 1 of 1
A Sun Yat-sen University student created BLA, an open-source tool that injects false clues into code artifacts to mislead AI reverse-engineering agents like REA, forcing them to waste up to 2800x more tokens on wild-goose chases.
AI securityLLM agentsadversarial attacks
0 likes · 8 min read

Shanghai Jiao Tong University researchers reveal 'Agent Resource Hijacking' attacks where attackers trick AI agents into consuming enterprise GPU compute and API quotas without stealing credentials, achieving 70-89% success rates across top models like GPT-5.5 and DeepSeek-V4-Pro, while proposed ResGate defense reduces success to 23.6% with minimal usability impact.
AI safetyAI securityLLM agents
0 likes · 16 min read

This review examines how evolutionary computation supports trustworthy AI across three research lines: evolutionary attacks to find failures, evolutionary defenses to search protective measures, and trustworthy self-evolution to govern which changes persist long-term, highlighting the shift from static robustness to continuous governance of memory, skills, and safeguards.
AI safety benchmarksMulti-Objective OptimizationTrustworthy AI
0 likes · 31 min read

This survey examines proactive protection methods—ranging from privacy filters and non‑learnable samples to generative safeguards, adversarial captchas, and traceability mechanisms—that embed adversarial perturbations before visual content is shared, trained, generated, accessed, or disputed, and evaluates them across transferability, adaptability, and deployment maturity.
AI securityadversarial attacksadversarial captcha
0 likes · 19 min read

The paper reveals that prompt‑compression modules, widely used to reduce token usage in LLM agents, can be exploited to erase critical safety constraints, enabling a black‑box attack called COMA that achieves up to 71% success across multiple compressors and tasks, and proposes isolated compression as an effective defense.
Agentic AICOMALLM security
0 likes · 10 min read

A new study presented at ICML 2026 reveals that the fuzzy matching used in LLM semantic caching creates an integrity vulnerability, allowing attackers to craft adversarial suffixes that cause cache‑key collisions and achieve up to 86 % response‑hijacking success on major cloud services such as AWS and Azure.
AI agentsLLM securityadversarial attacks
0 likes · 9 min read

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.
AI age verificationComputer VisionDiscord
0 likes · 6 min read

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.
AI securityEmbodied AIadversarial attacks
0 likes · 26 min read

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.
CNNTransformeradversarial attacks
0 likes · 14 min read

The article reviews Andrew Ng’s ‘AI for Everyone’ course, explaining AI bias with a facial‑recognition example and mitigation tips, describing adversarial attacks and defenses, outlining harmful AI uses such as deepfakes and privacy threats, and discussing McKinsey’s job‑displacement and creation forecasts alongside AI‑assisted programming tools.
AI biasAI ethicsAI tools
0 likes · 5 min read

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.
ImageNetadversarial attacksdeep learning security
0 likes · 8 min read

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.
ImageNetadversarial attacksdeep learning robustness
0 likes · 7 min read

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.
AI robustnessLLM securityReward Modeling
0 likes · 10 min read

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.
AI securityBlockchain SecurityGenerative AI
0 likes · 10 min read


AntTechDec 5, 2022 · Artificial Intelligence
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.
AAAI-23Differential PrivacyGraph Neural Networks
0 likes · 18 min read

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.
3D groundingCVPR 2022Computer Vision
0 likes · 15 min read

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.
Anti-FraudCAPTCHATraffic Classification
0 likes · 15 min read
