PrivacyEvalClaw: A Tool-Augmented Deliberative Agent for Auditing Privacy Leakage in LLM Jailbreak Responses

Reliable evaluation is a critical bottleneck in privacy red-teaming of large language models (LLMs): determining whether

历史摘要 · 未重新核验 · arXiv cs.CR

What Victims Take Away: Attack Reuse and Defense Revision after Prompt-Injection Attacks

Attack logs and red-team reports show targets the attacks used against them, to help them repair their defenses. Prompt

历史摘要 · 未重新核验 · arXiv cs.CR

Detecting and Mitigating Query-Based System Prompt Leakage Attacks

Large Language Models (LLMs) rely on proprietary system prompts to enforce task directives, safety policies, and access

历史摘要 · 未重新核验 · arXiv cs.CR

BUSTA: Teaching Small Language Models to Reason About Binary Execution

Large Language Models (LLM) are increasingly applied to Reverse Engineering (RE) workflows. RE, however, differs from co

历史摘要 · 未重新核验 · arXiv cs.CR

Bringing Hardware Security Education to All: A Seamless Curriculum Integration Approach

Cybersecurity education has predominantly focused on software and network security, under the assumption that the underl

历史摘要 · 未重新核验 · arXiv cs.CR

Log-To-Leak: Prompt Injection Attacks on Tool-Using LLM Agents via the Model Context Protocol

LLM agents increasingly access external tools through the Model Context Protocol (MCP), but the natural-language metadat

历史摘要 · 未重新核验 · arXiv cs.CR

Reva: Real-Time Vulnerability Detection through Contextual Probing of Large Language Models

A vulnerability costs more to fix the later it is discovered. As coding agents generate code faster than developers can

历史摘要 · 未重新核验 · arXiv cs.CR

Adversary Brains: Hypothesis-Based Logical Signatures for Threat Actor Detection with LLMs

An adversary brain is an LLM-based model of a threat actor's behavior, held as an evolving logical signature: a hypothes

历史摘要 · 未重新核验 · arXiv cs.CR

Then there is prompt injection, a malicious email, ...: The Wild Wide West of OpenClaw Agentic System Security

AI-powered personal assistant frameworks such as OpenClaw combine LLM reasoning, persistent memory, tool invocation, loc

历史摘要 · 未重新核验 · arXiv cs.CR

Can Open Be Too Open? On the Security and Trustworthiness of Fully Transparent LLMs

There are different levels of transparency for LLMs: open weights, open data, and/or open pipeline. Recently, more inter

历史摘要 · 未重新核验 · arXiv cs.CR

GuardRAG: A Post-Retrieval Defense against Poisoning Attacks in Retrieval-Augmented Generation

Retrieval-Augmented Generation (RAG) enhances the factual grounding of Large Language Models (LLMs) by incorporating ext

历史摘要 · 未重新核验 · arXiv cs.CR

Steering for Safety, Hijacking for Profit: Trojan Threats in the LLM Activation Steering Supply Chain

As large language models (LLMs) are increasingly integrated into Web search, online recommendation, and intelligent deci

历史摘要 · 未重新核验 · arXiv cs.CR

The Target Agent Is Its Own Best Critic but a Poor Proposer: Self-Improved Red Teaming

Automatic red teaming increasingly relies on large language models (LLMs) to iteratively diagnose failed attacks and pro

历史摘要 · 未重新核验 · arXiv cs.CR

Scope-Gap Injection: Exploiting Evaluator Channel Blind Spots in the OpenClaw Agent Security Competition

This paper introduces Scope-Gap Injection (SGI), an indirect prompt injection method for multi-stage LLM-agent safety pi

历史摘要 · 未重新核验 · arXiv cs.CR

From Representations to Computational Trajectories: Intrinsic Fingerprinting for Large Language Model Provenance Verification

Large language models (LLMs) are valuable digital assets, yet their weights can be easily modified after unauthorized ac

历史摘要 · 未重新核验 · arXiv cs.CR

SentryBench: A Cost-Sensitive Benchmark for Adversarial Robustness of Tool-Using LLM Agents in Public-Safety Deployments

Large language model (LLM) agents are increasingly wired to consequential tools that send public alerts, dispatch respon

历史摘要 · 未重新核验 · arXiv cs.CR

SymLift: Neural Semantic Lifting for Compile-Free Vulnerability Detection in Context-Incomplete Code Fragments

Missing project context can prevent build-dependent analyzers from constructing program relations required for vulnerabi

历史摘要 · 未重新核验 · arXiv cs.CR

Repairing Broken Access Control in Web Applications with LLMs: An Empirical Benchmark Study

Broken access control (BAC) ranks among the most severe web-application vulnerability classes, yet its automated repair

历史摘要 · 未重新核验 · arXiv cs.CR

CVE-2026-6307：漏掉一次签名比较，Wasm 返回类型错乱

CVE-2026-6307

CVE-2026-6307：漏掉一次签名比较，Wasm 返回类型错乱

来源摘录 · 先知社区

CVE-2026-39821：Go idna 纯 ASCII 标签解码漏报

CVE-2026-39821

CVE-2026-39821：Go idna 纯 ASCII 标签解码漏报

来源摘录 · 先知社区