arXiv:2611.18908v1 Announce Type: new Abstract: Alignment-trained large language model (LLM) agents are increasingly deployed in autonomous, multi-round settings where ethical compliance can come into direct tension with task success. While safety literature extensively document…
⏱ Độ trễ thật của tab này: trung bình 6.8 giờ sau khi CVE được công bố (n = 5426, sớm nhất 0.1 giờ). Số âm nghĩa là ta thấy trước ngày công bố.
arXiv:2611.17820v1 Announce Type: new Abstract: Previous work has shown that alignment training on individual LLM agents can fail to transfer to multi-agent settings. We investigate the underlying mechanisms for this failure, and explore interventions to preserve alignment trans…
arXiv:2611.22971v1 Announce Type: new Abstract: Safety benchmarks for computer-use agents (CUAs) place the harm either in the intent of the task or on a single screen: a malicious instruction, an injected command, a phishing page. An agent that checks the task and each screen in…
arXiv:2611.19237v1 Announce Type: new Abstract: Human approval provides effective control only if rejected work stops and authorized work completes. We formalize refusal safety and bounded authorization completeness, and audit three execution boundaries: the model selects a call…
arXiv:2611.22583v1 Announce Type: new Abstract: Tool-using language-model agents must act on information from untrusted emails, webpages, and documents. This creates a central security challenge: malicious instructions and legitimate task directives can arrive through the same i…
arXiv:2611.18883v1 Announce Type: new Abstract: Natural language optimization (NLO) improves an agent by having a model critique and rewrite its natural language policy parameters: system prompts, tool instructions, reasoning directives, few-shot demonstrations. These artifacts …
arXiv:2611.19915v1 Announce Type: new Abstract: Memory-safety vulnerabilities continue to dominate real-world exploit chains despite decades of research into software mitigations. CHERI is a hardware architecture that enforces spatial and temporal memory safety by augmenting poi…
arXiv:2611.14864v1 Announce Type: new Abstract: An AI agent's decision not to act can itself become an attack surface. We study warning-induced over-abstention, an availability attack in which adversaries place fabricated risk warnings in low-trust content to prevent legitimate …
arXiv:2611.24682v1 Announce Type: new Abstract: Language models judging commands to industrial controllers may not receive machine state, the equipment's current condition. The same command can require opposite safety decisions in different states, so an incorrect answer may ref…
arXiv:2611.17818v1 Announce Type: new Abstract: Jailbreak attacks can circumvent the safety mechanisms of Large Language Models (LLMs) and lead to the generation of harmful or policy-violating content. We first examine eight jailbreak methods against an LLM-based filter and obse…
Abstention, the withholding of a response by a large language model (LLM) under uncertainty or safety risks, is essential for trustworthy models. The causes of abstention fall into three perspectives: ambiguity in the Query (Q), limits of Model Knowledge (MK), and misalignment wi…
arXiv:2611.16565v1 Announce Type: new Abstract: We ask whether small, open-weight decision models can support vulnerability discovery in open-source software at scale and what the surrounding system must provide. These models answer a typed question in one forward pass, returnin…
arXiv:2611.21682v1 Announce Type: new Abstract: Large language models (LLMs) are increasingly being explored for knowledge-intensive and safety-critical applications in nuclear engineering. However, existing nuclear-domain benchmarks provide limited content coverage or focus pri…
arXiv:2611.21832v1 Announce Type: new Abstract: As AI systems are increasingly used to automate AI R&D, misaligned systems could sabotage research intended to improve model safety and alignment. To stress-test defenses against such sabotage, we use a human expert as a proxy sabo…
arXiv:2611.22069v1 Announce Type: new Abstract: Safety is personal: the same action can be safe for one person but unsafe for another. Yet current guardrails judge content without considering who is affected. Using medication safety, we introduce \textit{patient-conditioned medi…
arXiv:2611.22186v1 Announce Type: new Abstract: Chat models on phones and laptops are usually 4-bit copies of the model that was evaluated, and audits of such copies disagree about what quantization does to refusal. A 4-bit build is one of many weight perturbations of the same s…
arXiv:2611.22292v1 Announce Type: new Abstract: Multimodal large language models (MLLMs) are rapidly advancing, yet their safety remains a critical concern, especially under jailbreak attempts. Existing MLLM jailbreaks either inject fragile adversarial perturbations into irrelev…
arXiv:2611.22773v1 Announce Type: new Abstract: Swarms of unmanned aerial and ground vehicles (UAV/UGV) operating in contested environments cannot rely on software alone for physical safety, since the software stack and its networks are part of the attack surface. The Guarded Sw…
Personalized vision-language assistants use visual observations and personal memories to provide advice tailored to the user. When the observed scene contains a physical hazard, personal context, such as allergies, medical conditions, or mobility constraints, can change what help…
arXiv:2611.18930v1 Announce Type: new Abstract: Safety benchmarks are cited as the primary tool to understand the risks posed by generative AI (GenAI) systems, yet growing evidence suggests they often fail to meet the needs of real-world safety evaluation. We present findings fr…
arXiv:2611.18971v1 Announce Type: new Abstract: Child safety in AI depends on the accurate identification of minor users. Young children tend to talk rather than type when interacting with AI, so in practice this identification requires predicting age from speech. We evaluated w…
arXiv:2611.19166v1 Announce Type: new Abstract: Every medication comes with two types of medical facts: indications, which describe what conditions the drug treats (benefits), and contraindications, which describe who must avoid it to prevent serious harm (safety limits). Althou…
arXiv:2611.18910v1 Announce Type: new Abstract: Memory safety rules out classical code reuse attacks that rely on arbitrary memory corruption, but does not eliminate code reuse altogether. Attacks such as object injection, prototype pollution, class pollution, DOM clobbering, an…
arXiv:2611.19090v1 Announce Type: new Abstract: As Large Language Models (LLMs) become used more commonly across a variety of fields, a growing body of research has highlighted particular concerns surrounding model security. Notably, prompt injection attacks, where malicious ins…
arXiv:2611.20459v1 Announce Type: new Abstract: AI models can exhibit deceptive behaviors such as strategically underperforming on evaluations (sandbagging), faking alignment, or covertly acting on behalf of a hidden principal. Safe deployment therefore requires auditing techniq…
arXiv:2611.17620v1 Announce Type: new Abstract: High-level behaviors such as sycophancy and hallucination correspond to directions in a model's activation space (Chen et al., 2025). These directions, known as persona vectors, can suppress undesirable traits either at inference t…
arXiv:2611.18055v1 Announce Type: cross Abstract: DeCompCert is a Rocq-verified decompiler for AArch64 assembly generated by CompCert. It reconstructs CompCert C through nine reverse passes, combining deductively verified transformations with translation validation. Untrusted OC…
arXiv:2611.18563v1 Announce Type: cross Abstract: Monitoring maritime activities is important for safety and security, particularly for detecting illegal activities, but remains challenging due to the size and complexity of the maritime domain. The Automatic Identification Syste…
arXiv:2611.16151v1 Announce Type: new Abstract: Hardware fault injection attacks threaten machine learning systems, yet their impact on reinforcement learning (RL) has received little attention despite the use of RL in safety-critical applications. This paper presents a security…