🧩 Chức năng: Công cụ lập trình (xem mô tả gốc bên dưới)
Thư viện tensor cốt lõi cho các dự án machine learning hiệu năng cao, đặc biệt tối ưu cho việc chạy mô hình AI trên CPU. Đây là nền tảng đứng sau nhiều dự án LLM nổi tiếng như llama.cpp.
⏱ Độ trễ thật của tab này: trung bình 7.1 giờ sau khi CVE được công bố (n = 5350, sớm nhất 0.1 giờ). Số âm nghĩa là ta thấy trước ngày công bố.
🧩 Chức năng: Công cụ lập trình (xem mô tả gốc bên dưới)
Thư viện tensor cốt lõi cho các dự án machine learning hiệu năng cao, đặc biệt tối ưu cho việc chạy mô hình AI trên CPU. Đây là nền tảng đứng sau nhiều dự án LLM nổi tiếng như llama.cpp.
🧩 Chức năng: Hậu kỳ & ghép (ComfyUI, upscale, nội suy khung, phụ đề)
Bộ node tùy chỉnh giúp đơn giản hóa và tích hợp các chức năng thường dùng trong ComfyUI, hỗ trợ tối ưu quy trình làm việc với Stable Diffusion. Rất hữu ích cho người dùng muốn tinh giản giao diện và tăng hiệu suất tạo ảnh.
arXiv:2611.17883v1 Announce Type: new Abstract: Multi-agent systems solve complex tasks by decomposing them into subtasks, delegating the subtasks to agents, and aggregating the agents' responses using an aggregation rule. Malicious agents can manipulate the final solution by pr…
arXiv:2611.10756v1 Announce Type: new Abstract: Large Language Models (LLMs) are increasingly deployed as shared multi-tenant services, relying on advanced optimizations like KV cache and prefix sharing to improve throughput and reduce latency. However, these optimizations intro…
arXiv:2611.17621v1 Announce Type: new Abstract: Kernel optimization is essential for efficient large language model (LLM) training and inference on Ascend NPUs. Existing Ascend kernel benchmarks primarily focus on kernel generation, with less emphasis on the ability of agents to…
Weight-space programming, introduced in recent MIT work by Takehi et al., enables a frontier LLM agent to read a network's weights and write a repair. Their GPT-6 Astra agent corrected the reasoning loops of Qwen3.5-0.8B without gradients, using 158 agent turns, 89.7 million toke…
arXiv:2611.13015v1 Announce Type: new Abstract: LLM agents are increasingly asked to make plans for people, such as scheduling drivers or assigning nurses. However, people's requests are often incomplete, and the supporting information for decision-making may be outdated, requir…
Where does answer engine optimization (AEO) end and agent experience begin? Inside Resend, from the docs that help Claude Code pick an email API to the tools it uses to send.
arXiv:2611.10542v1 Announce Type: new Abstract: Large language model (LLM) agents increasingly improve themselves through self-evolution, distilling execution experience into persistent, reusable skills. However, this persistence creates a new attack surface: malicious rules imp…
10.0.14393.0, 10.0.17763.0, 10.0.19044.0, 10.0.19045.0, 10.0.22631.0, 10.0.22631.0 +4
CVE-2026-84004 tương tự CVE-2026-84005, khai thác cùng cơ chế tối ưu hoá sai trong AMD Zen để tiết lộ thông tin nội bộ cho kẻ tấn công đã có quyền truy cập.
10.0.17763.0, 10.0.22631.0, 10.0.22631.0, 10.0.26100.0, 10.0.26200.0, 10.0.28000.0 +4
CVE-2026-84005 là lỗ hổng Information Disclosure trong bộ xử lý AMD Zen, do tối ưu hoá loại bỏ hoặc sửa đổi mã bảo mật quan trọng, cho phép kẻ tấn công tiết lộ thông tin nội bộ trên máy.
🧩 Chức năng: Khung/chương trình AI agent tự làm việc nhiều bước
Công cụ mới từ shadcn giúp tối ưu hóa chi phí phát triển bằng cách sử dụng các mô hình AI mạnh mẽ (như GPT-4/Claude 3.5) để audit mã nguồn, sau đó lập kế hoạch chi tiết cho các mô hình rẻ hơn thực hiện tác vụ. Giải pháp này rất hữu ích để tự động hóa quy trình cải tiến code một cách tiết kiệm.
arXiv:2611.14681v1 Announce Type: new Abstract: Today's AI workloads are rapidly changing: new models are released more often, model architectures are increasingly diverse, and traffic distributions change frequently. The challenge we face at Together AI is continual optimizatio…
arXiv:2611.17722v1 Announce Type: new Abstract: Multi-level learning/optimization, e.g., bilevel optimization, has been recognized for its versatile applicability across a broad array of domains, spanning from signal processing to machine learning. However, most existing works f…
arXiv:2611.17393v1 Announce Type: new Abstract: End-to-end autoencoder communication has emerged as a promising means of achieving efficient data transmission across various domains. Nevertheless, vulnerabilities have recently been revealed in these systems by adversarial attack…
arXiv:2603.10099v2 Announce Type: replace-cross Abstract: The 2020 United States Census adopted differential privacy to protect individual confidentiality, adding calibrated noise to summary demographic measurements that are then reconciled into population tables. We developed B…
18, 17, 16, 15, 0
Type confusion in PostgreSQL min/max aggregate optimization allows an object creator to execute arbitrary code as the operating system user running the database, via a user-defined aggregate whose SORTOP option causes the planner to treat the aggregate's input and result as the s…
arXiv:2611.13942v1 Announce Type: new Abstract: Cyberattacks against power systems increasingly exploit industrial communication and control dependencies arising from distributed energy resources (DER), remote management, and market-based incentives. Progress on intrusion detect…
arXiv:2611.14924v1 Announce Type: new Abstract: Deployment of retrieval-augmented generation (RAG) for Edge-AI is constrained by memory and bandwidth limitations. A standard 7B parameter model consumes 14 GB Video Random Access Memory (VRAM) in half-precision (FP16), precluding …
arXiv:2611.15238v1 Announce Type: new Abstract: Discrete diffusion generation is emerging as a competitive alternative to autoregressive generation by denoising the entire token sequence through iterative refinement. Its global planning and progressive correction capabilities ar…
arXiv:2611.15389v1 Announce Type: new Abstract: Titan is a compelling target for future deep space exploration, while its extreme Earth-Titan distance and complex environment make continuous ground-in-the-loop planning impractical. Existing planning methods, primarily developed …
0
The Smush WordPress plugin before 4.3.6 does not verify that a user is authorised to act on the targeted media attachment, only checking a broad upload-files capability, which allows users with the Author role and above to optimise and restore attachments owned by other users, a…
arXiv:2512.08289v5 Announce Type: replace Abstract: Retrieval-Augmented Generation (RAG) systems enhance LLMs with external knowledge but introduce a critical attack surface: corpus poisoning. While recent studies have demonstrated the potential of such attacks, they typically r…
arXiv:2611.15091v1 Announce Type: new Abstract: We present a GPU-CPU hybrid framework for Weighted Partial MaxSAT (WPMAX-SAT) that complements state-of-the-art CPU solvers with differentiable GPU optimization. The GPU stage directly optimizes instance-specific categorical decisi…
arXiv:2611.12954v1 Announce Type: new Abstract: Recent advances in large language models (LLMs) have enabled weight-space programming (WSP), which directly generates neural network weights without gradient-based optimization. However, extending WSP to physical problems remains c…
arXiv:2611.13236v1 Announce Type: new Abstract: We present the Neural Control Plane (NCP), a closed-loop system that observes language-model execution, synthesizes candidate weight-space intervention rules, evaluates them, and maintains a versioned table of admitted policies. NC…
arXiv:2611.10180v1 Announce Type: new Abstract: Medical ultrasound (US) enables real-time, non-invasive diagnosis, yet building US-based AI systems is limited by data scarcity, costly expert annotation, and limited cross-center generalizability. Privacy constraints and instance-…
arXiv:2611.10279v1 Announce Type: new Abstract: Safety-aligned language models are known to suffer degradation in refusal behavior following domain-specific fine-tuning, even when the fine-tuning data is entirely benign. The predominant mitigation strategy, preference optimizati…
arXiv:2611.10432v1 Announce Type: new Abstract: Speculative decoding efficiency depends on the number of consecutively accepted draft tokens, yet token-level training objectives do not directly optimize this quantity. We introduce D-REAL (Drafter Reinforcement of Expected Accept…
arXiv:2611.10707v1 Announce Type: new Abstract: Machine learning-based intrusion detection systems (IDSs) can improve IoT security but often rely on large feature sets that increase computational overhead and limit deployment on resource-constrained devices. This paper presents …
RLVR trains language models to reason by rewarding correct final answers. A widely used algorithm, group-relative policy optimization (GRPO), gives every token of a solution the same credit, so dense process rewards have been proposed to credit intermediate steps, for example wit…