cập nhật 41 phút trước

⏱ Độ trễ thật của tab này: trung bình 6.8 giờ sau khi CVE được công bố (n = 5439, sớm nhất 0.1 giờ). Số âm nghĩa là ta thấy trước ngày công bố.

23
Disentangling Robustness and Sycophancy in Multilingual LLMs
📄 Paper arXiv cs.AI 27 ngày trước

arXiv:2610.30696v1 Announce Type: new Abstract: When a multilingual LLM agrees with a user's false premise, is it being sycophantic, or does it simply not know better in that language? Current research does not distinguish these cases: they record only whether the model agreed, …

🅲 hạng C
23
Amplifying Latency and Cost of LLM Inference via Rowhammer Attack
📄 Paper arXiv cs.CR 27 ngày trước

arXiv:2610.30798v1 Announce Type: new Abstract: Large language models (LLMs) achieve high inference efficiency using optimized GPU kernels generated by just-in-time (JIT) compilation frameworks such as Triton. Triton compiles GPU kernels through an MLIR/LLVM-based pipeline. We i…

🅲 hạng C
23
Weaponizing the Exit: Input Privacy Leakage via Exit Routing Manipulation in Multi-Exit Networks
📄 Paper arXiv cs.CR 27 ngày trước

arXiv:2610.31246v1 Announce Type: new Abstract: Multi-exit networks improve inference efficiency by allowing inputs to terminate early at intermediate exits. We identify a privacy vulnerability in cloud deployments of these networks: a malicious Cloud Service Provider (CSP) can …

🅲 hạng C
👁 Trúng danh sách theo dõi: kw:supply chain
23
Automated screening of privacy risks in medical AI
📄 Paper arXiv cs.CR 27 ngày trước

arXiv:2610.30229v1 Announce Type: cross Abstract: Medical artificial intelligence (AI) models are often shared without their training data, assuming this protects the patients who contributed them. However, a model can reveal whether a patient's records were used in training, a …

🅲 hạng C
23
Training an MoE LLM in MXFP4: How We Closed the Gap to BF16
📄 Paper arXiv cs.AI 24 ngày trước

arXiv:2610.32229v1 Announce Type: new Abstract: Training large language models (LLMs) in sub-8-bit arithmetic promises substantial gains in memory bandwidth and compute throughput, yet FP4 formats introduce quantization error severe enough to destabilize optimization if left una…

🅲 hạng C
23
Fooling LLM Safety Guards into Judging the Wrong Text
📄 Paper arXiv cs.CR 24 ngày trước

arXiv:2610.31695v1 Announce Type: new Abstract: Safety guards check content before an LLM application acts on it, and most paste that content into a plain-text prompt. Content that imitates the prompt can append a second task, and the guard answers that one instead. We call this…

🅲 hạng C
23
SynMark: Semantic-Coordinate Multi-Bit Watermarking during LLM Code Generation
📄 Paper arXiv cs.CR 24 ngày trước

arXiv:2610.32017v1 Announce Type: new Abstract: LLM-generated code is increasingly reused across downstream applications and often undergoes identifier renaming or structural refactoring, making robustness to functionality-preserving edits essential for generative multi-bit code…

🅲 hạng C
23
Democratic Accountability for Autonomous Systems: A Legislative Notification Framework and Comparative Institutional Design
📄 Paper arXiv cs.CR 24 ngày trước

arXiv:2610.32226v1 Announce Type: cross Abstract: No domestic legislature has ever reviewed a state's decision to authorise an autonomous cyber weapon before it acted. Closing the international attribution gap is necessary but not sufficient for democratic accountability. We for…

🅲 hạng C
23
Spoofing the Toolbox: Optimizing Tool Descriptions to Deceive LLM Agents
📄 Paper arXiv cs.CR 23 ngày trước

arXiv:2610.33182v1 Announce Type: new Abstract: Large language model (LLM) agents are widely used in many applications and often rely on external tools. Tool information, such as function descriptions, helps agents determine which tools best suit a given context. However, attack…

🅲 hạng C
23
CVE-to-ATT&CK Mapping via Evidence-Constrained Ranking
📄 Paper arXiv cs.CR 23 ngày trước

arXiv:2610.33924v1 Announce Type: new Abstract: Mapping CVEs to MITRE ATT&CK techniques helps defenders translate vulnerability information into hypotheses about adversary behavior. Automating this mapping is challenging because relevant evidence is fragmented across heterogeneo…

🅲 hạng C
23
Can LLMs Uncover Hidden Physical Laws through Free Exploration?
📄 Paper HF Daily Papers 23 ngày trước

Scientific discovery often begins with an unexpected observation that prompts a new research question. In a world with altered physical laws, can large language models (LLMs) notice unexpected behavior, such as heavier objects falling more slowly, and discover the law through exp…

🅲 hạng C
23
Scoped Semantic Invariance as a Foundation for LLM Reliability
📄 Paper arXiv cs.AI 23 ngày trước

arXiv:2610.33267v1 Announce Type: new Abstract: LLMs often answer semantically equivalent queries inconsistently, even when the underlying facts are known. We formalize consistency as semantic invariance over equivalence classes of queries and show that invariance under arbitrar…

🅲 hạng C
23
How To Build A Modern Diffusion Language Model
📄 Paper arXiv cs.AI 23 ngày trước

arXiv:2610.33391v1 Announce Type: new Abstract: The frontier of large language model (LLM) performance has advanced rapidly in recent years, fueled by scaling laws for pretraining on large text corpora. However, mainstream LLMs have fundamental limitations: they generate text au…

🅲 hạng C
23
Skill Dreaming: Evolving Repository Skills through Pull Request Streams
📄 Paper arXiv cs.AI 22 ngày trước

arXiv:2610.36188v1 Announce Type: new Abstract: Large language model (LLM) agents increasingly rely on reusable skills for complex software engineering tasks, making skill libraries essential for repository development and maintenance. However, repository-specific skills must ev…

🅲 hạng C
23
Verifiable Efficiency Limits of Innate LLM Planning in Sliding Puzzles
📄 Paper arXiv cs.AI 22 ngày trước

arXiv:2610.36415v1 Announce Type: new Abstract: Tool-assisted planning success can reflect external search and verification, leaving a model's unaided planning ability unresolved. We isolate innate planning through complete plans generated in one response without tools or execut…

🅲 hạng C
23
PHP2Shell and the Future of AI-Discovered Vulnerability Disclosure: Evidence, Verification, and Human Authority
📄 Paper arXiv cs.CR 22 ngày trước

arXiv:2610.36334v1 Announce Type: new Abstract: AI-assisted vulnerability discovery can produce technically plausible findings faster than maintainers can validate, coordinate, and safely disclose them. The resulting challenge is not only to establish whether a vulnerability cla…

🅲 hạng C
23
GKS Fully Homomorphic Encryption: Security Analysis
📄 Paper arXiv cs.CR 22 ngày trước

arXiv:2610.37061v1 Announce Type: new Abstract: All traditional encryption schemes share a fundamental operational vulnerability: encrypted data must be decrypted before it can be used. Fully homomorphic encryption eliminates this vulnerability by enabling computation directly o…

🅲 hạng C
23
Revisiting Privacy Risks in Quantized Deep Neural Networks
📄 Paper arXiv cs.CR 22 ngày trước

arXiv:2610.37488v1 Announce Type: new Abstract: Model quantization has become a fundamental technique for enabling the efficient deployment of large-scale deep neural networks (DNNs) in data-intensive applications by reducing model size. Despite its widespread adoption, the impa…

🅲 hạng C