81 matches found
TRACE
TRACE: Task-Aware Adaptive Self-Evolving Agentic Jailbreaking Official implementation for TRACE: Task-Aware Adaptive Self-Evolving Agentic Jailbreaking. 📖 Overview Large language model LLM agents can autonomously plan, invoke tools, interact with environments, and execute long-horizon workflows...
Astra Linux – Vulnerability in openjpeg2
A flaw was discovered in OpenJPEG’s encoder, specifically in the opjdwtcalcexplicitstepsizes function. This flaw allows an attacker who can provide crafted inputs for the decomposition levels to cause a buffer overflow. The greatest threat of this vulnerability is to system availability...
Towards a Unified Misuse Monitoring Benchmark
LLM agents increasingly act in multi-actor environments, exposing them to misuse from multiple sources: decomposition attacks, where a harmful request is split into innocuous sub-requests, and prompt injection attacks, where a compromised tool delivers a malicious instruction. Existing evaluation...
Friday Squid Blogging: Rotting Squid on a Beached California Boat
Smells awful: But an estimated 30 to 50 tons of dead squid remain inside the boat's catch tank, where they have been decomposing for days. "That is nasty. I wouldn't want to do that," said commercial fisherman Dick Ogg of the Bodega Bay Fishermen's Marketing Association. Ogg said anyone familiar...
A Controlled Candidate-Set Benchmark for Offline Satellite-Security Plan Decomposition
Some security-development settings require local models that map an objective to an ordered, checkable plan. We present a low-rank decomposition adapter and release a case-disjoint corpus with 24 authored decompositions and 83 derived next-step examples across 24 satellite-security cases. To...
Hybrid Latent-Structural Fusion (HLSF) for Cyber Anomaly Detection
Malicious anomalous activity detection is a fundamental challenge for cyber security systems. Both tensor decomposition under statistical framework with CANDECOMP-PARAFAC alternating Poisson regression CP-APR and normalizing flows have proven to be powerful unsupervised machine learning methods...
OpenAnt: LLM-Powered Vulnerability Discovery through Code Decomposition, Adversarial Verification, and Dynamic Testing
Automated vulnerability discovery in large codebases remains challenging: traditional static analysis produces high false-positive rates, while dynamic approaches such as fuzzing require substantial infrastructure and often target narrow classes of bugs. Recent advances in large language models...
JLSEC-2026-537
A flaw was found in OpenJPEG’s encoder in the opjdwtcalcexplicitstepsizes function. This flaw allows an attacker who can supply crafted input to decomposition levels to cause a buffer overflow. The highest threat from this vulnerability is to system availability...
PT-2026-47102
A flaw was found in OpenJPEG’s encoder in the opj dwt calc explicit stepsizes function. This flaw allows an attacker who can supply crafted input to decomposition levels to cause a buffer overflow. The highest threat from this vulnerability is to system availability...
Disentangling Adversarial Prompts: A Semantic-Graph Defense for Robust LLM Security
Large Language Models LLMs are increasingly vulnerable to adversarial prompts that exploit semantic ambiguities to bypass safety mechanisms, resulting in harmful or inappropriate outputs. Such attacks, including jailbreaking and prompt injection, pose significant risks to the integrity and...
SEED: Semi-Supervised Continual MalwarE Detection for Tackling ConcEpt Drift on a BuDget
Machine learning based malware detectors become obsolete over time due to concept drift in benign and malware applications. Recent methods rely on fully labeled data and use hierarchical contrastive loss HCL with active learning to improve robustness against drift by exploiting semantic structure...
Reasoning As an Attack Surface: Adaptive Evolutionary CoT Jailbreaks for LLMs
Large Reasoning Models LRMs have demonstrated remarkable capabilities in reasoning and generation tasks and are increasingly deployed in real-world applications. However, their explicit chain-of-thought CoT mechanism introduces new security risks, making them particularly vulnerable to jailbreak...
Do Coding Agents Understand Least-Privilege Authorization?
As coding agents gain access to shells, repositories, and user files, least-privilege authorization becomes a prerequisite for safe deployment: an agent should receive enough authority to complete the task, without unnecessary authority that exposes sensitive surfaces.To study whether current...
Spring AI Agentic Patterns (Part 3): Why Your AI Agent Forgets Tasks (And How to Fix It)
Have you ever asked an AI agent to perform a complex multi-step task, only to find it skipped a critical step halfway through? You're not alone. Research shows that LLMs struggle with "lost in the middle" failures—forgetting tasks buried in long contexts. When your agent juggles file edits, test...
CVE-2026-22705
CVE-2026-22705 — RustCrypto: Signatures (ML-DSA) exhibits a timing side-channel in the Decompose algorithm used during signing. The vulnerability arises from variable-time division of r1 by TwoGamma2::U32 on secret-derived data, leaking information about the signing key through timing variations....
A Wolf in Sheep's Clothing: Bypassing Commercial LLM Guardrails Via Harmless Prompt Weaving and Adaptive Tree Search
Large language models LLMs remain vulnerable to jailbreak attacks that bypass safety guardrails to elicit harmful outputs. Existing approaches overwhelmingly operate within the prompt-optimization paradigm: whether through traditional algorithmic search or recent agent-based workflows, the...
OpenVM Buffer Overflow Registry
The vulnerability of the OpenVM platform is related to incorrect calculations of the size of the buffer allocated in byte-based decomposition of pc on the AUIPC chip. Exploiting this vulnerability can allow a remote attacker to force the destination register to accept a different value...
DrAttack
DrAttack: Prompt Decomposition and Reconstruction Makes Powerf...
