cập nhật 25 phút trước

⏱ 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.

28
Aligning AI Teams
📄 Paper arXiv cs.AI 3 ngày trước

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…

🅲 hạng C
26
Auditable Authorization for LLM Agents: Empirical Evidence Across Model, Harness, and Workflow Layers
📄 Paper arXiv cs.AI 2 ngày trước

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…

🅲 hạng C
25
SPLICE: Structured Prompt Local Iterative Combinatorial Evolution
📄 Paper arXiv cs.AI 2 ngày trước

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 …

🅲 hạng C
24
The Limit Is in the Input, Not the Model: State-Dependent Command Safety in Natural-Language Interfaces to Industrial Controllers
📄 Paper arXiv cs.CR 14 giờ trước

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…

🅲 hạng C
23
LLMs are Vulnerable to Malicious Requests with Indistinguishable Intent
📄 Paper arXiv cs.CR 3 ngày trước

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…

🅲 hạng C
23
Abstention Direction in Large Language Models: Mechanistic Interpretability of Abstention Across Query Ambiguity, Knowledge Limits, and Safety Alignment
📄 Paper HF Daily Papers 4 ngày trước

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…

🅲 hạng C
23
Look Here, Then Decide: Small Decision Models in a Harness Find Memory-Safety Vulnerabilities at Scale on Commodity GPUs
📄 Paper arXiv cs.CR 6 ngày trước

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…

🅲 hạng C
22
NEBench R1: a Reasoning Benchmark for Evaluating Nuclear Domain Competence in Large Language Models
📄 Paper arXiv cs.AI 1 ngày trước

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…

🅲 hạng C
22
Auditors Struggle to Catch Sabotage In End-to-End Transcripts of AI Safety Research
📄 Paper arXiv cs.AI 1 ngày trước

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…

🅲 hạng C
22
Is 4-Bit Quantization More Than Noise to a Chat Model's Safety Behaviour? Attributing Behavioural Shifts with Matched Perturbations
📄 Paper arXiv cs.CR 1 ngày trước

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…

🅲 hạng C
22
PReMind: When Personal Memory Meets Visual Hazards
📄 Paper HF Daily Papers 2 ngày trước

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…

🅲 hạng C
21
Memory Safety Isn't All You Need: Code Reuse Attacks in Memory Safe Languages
📄 Paper arXiv cs.CR 2 ngày trước

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…

🅲 hạng C
21
When the Spec Sheet Lies: Injection Attacks on LLM-Generated Compliance Artifacts
📄 Paper arXiv cs.CR 2 ngày trước

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…

🅲 hạng C
21
Designing Deceptive Model Organisms that Successfully Evade Auditing
📄 Paper arXiv cs.CR 2 ngày trước

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…

🅲 hạng C
21
Safety Side Effects of Preventative Persona Steering: An Empirical Audit
📄 Paper arXiv cs.AI 3 ngày trước

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…

🅲 hạng C
21
DeCompCert: A Verified Decompiler for CompCert-Generated AArch64
📄 Paper arXiv cs.CR 3 ngày trước

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…

🅲 hạng C
21
Pairing AIS with RF: Analysis of a Markovian Approach for Maritime Data Fusion
📄 Paper arXiv cs.CR 3 ngày trước

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…

🅲 hạng C