TranScope: What the Software Hides About LLM Training Data, the Hardware Reveals at Scale, and Accelerators Magnify
Membership is the root privacy primitive in machine learning: to date, no hardware-based out-of-distribution detection o
2026-10-07 归档
今日共收录 1979 条安全情报，包含 1214 个 CVE，387 篇安全通告，以及 100 篇研究论文。
← 返回归档列表Jailbreaking Open-Weight LLMs via Random Embedding Perturbations
While open-weight models have enjoyed steady progress in capabilities and wide adoption across multiple domains, their s
Does AI Help Cyber Attackers or Defenders? Evidence from Nonpublic Vulnerabilities and Subsequent Attacks
The release decision for frontier AI systems increasingly relies on cyber capability benchmarks, yet public vulnerabilit
An Evaluation of the Semantic Understanding Capabilities of Large Language Models for Web Attack Payloads
Computer vision services delivered through Web interfaces and APIs process textual requests for image-resource acquisiti
RAISED: Self-Distillation for Robustness to Prompt Injection in LLM Agents
Tool-using language-model agents are vulnerable to indirect prompt injection because they must act on untrusted external
Correct Verdicts, Flawed Reasoning: Structured Auditing of LLM-based Vulnerability Reasoning
Large Language Models (LLMs) are increasingly deployed for automated software vulnerability analysis. Binary classificat
Towards a Unified Misuse Monitoring Benchmark
LLM agents increasingly act in multi-actor environments, exposing them to misuse from multiple sources: decomposition at
DP-ES: Differentially Private Evolution Strategies for Prompt Optimization
Token-level differentially private (DP) prompt optimization methods such as DP-OPT can become unstable under tight priva
Where Did the Repair First Go Wrong? Localizing the Origins of Silent Failures in Agentic Vulnerability Repair
Localizing where an LLM-based agent first fails to uphold security during a repair can show which stage of its workflow
Cross-Lingual Transferability of Training Data Extraction Attacks to Recover Memorized PII
The robustness of Personally Identifiable Information (PII) protection in Large Language Models (LLMs) is a critical con
Backdooring Sparse Autoencoders
Sparse autoencoders (SAEs) are increasingly used not only to interpret language models but also to intervene on their in
H-CRSPV: Preventing Semantic Omission in Late-Bound Large Language Model Releases
Large-language-model release pipelines increasingly combine commitments, signatures, provenance records, and heterogeneo
Compromise Is Not Consequence: Evaluating Task-Scoped Authorization in LLM Agents with Paired Replay
A tool-using model can follow a malicious instruction even when its credentials are valid. We study whether task-scoped
AgentDoxx: Agentic Re-identification of Anonymized Text with Web Search
As Large Language Models (LLMs) gain tool use capabilities such as web search, they can retrieve and cross-reference pub
Vyper: Memory corruption using function calls within tuples / nested calls
### Impact When performing a function call inside a tuple or as an argument inside another function call, there is a mem
Vyper: Call stack corruption when passing complex type containing non-base type members as argument
### Impact When we pass a multi-dimensional array (like `[[1, 2], [3, 4]]`) as an argument to internal/external function
Vyper: Return inside for loop more than 1 level deep
# VVE-2020-0002 Earlier today, we received a responsible disclosure of a potential issue from @michwill (developer of @c
[原创]某游戏大厂反作弊驱动分析：从代码去混淆到敏感行为还原，CVE-2025-45737任意内核读写漏洞复现、提权实验，附PoC
CVE-2025-45737
[原创]某游戏大厂反作弊驱动分析：从代码去混淆到敏感行为还原，CVE-2025-45737任意内核读写漏洞复现、提权实验，附PoC
Critical Atlassian Flaw Lets Unauthenticated Attackers Read Known Files Across 8 Products
CVE-2026-21589
A critical flaw in 8 Atlassian Data Center products, which customers host themselves, allows an attacker with no login a
Mission-Aware Attestation Envelopes for Time-Critical Autonomous Action: A Hardware-in-the-Loop V2I Study
An autonomous system that asks for a privileged physical action is usually gated on integrity evidence: a platform prove