cập nhật 1 giờ trước

⏱ Độ trễ thật của tab này: trung bình 7.2 giờ sau khi CVE được công bố (n = 5386, sớm nhất 0.1 giờ). Số âm nghĩa là ta thấy trước ngày công bố.

32
I2C-GRPO: Insight Importance Checklist for Reinforcement Learning in a Multi-Agent DDoS Alert Triage Workflow
📄 Paper arXiv cs.CR 13 giờ trước

arXiv:2611.17534v1 Announce Type: new Abstract: Large language model (LLM) agents are increasingly used for investigative tasks over incomplete evidence, such as cybersecurity alert triage, but reinforcement learning (RL) for such tasks is difficult: intermediate steps have no v…

🅲 hạng C
32
Cyber-Zero: Training Cybersecurity Agents without Runtime
📄 Paper arXiv cs.CR 13 giờ trước

arXiv:2508.00910v3 Announce Type: replace Abstract: Large Language Models (LLMs) have achieved remarkable success in software engineering tasks when trained with executable runtime environments, particularly in resolving GitHub issues. However, such runtime environments are ofte…

🅲 hạng C
28
datasette-agent 0.6a0
🤖 AI Agent / MCP Simon Willison 1 ngày trước

Release: datasette-agent 0.6a0 Accumulated small fixes for Datasette Agent, my agent plugin for Datasette that lets you ask questions about your data. Most notably, it now adds an agent prompt directly to the Datasette homepage, with support for hash-mentions of tables and querie…

🅲 hạng C
28
Experience-Handoff Reinforcement Learning for Recursive Self-Improvement
📄 Paper HF Daily Papers 4 ngày trước

It is common to train Large Language Models (LLMs) for agentic tasks with Reinforcement Learning (RL) from scalar rewards. However, this throws away the fine-grained information about the environment that the agent experienced during the trial (e.g., error logs). To utilize this …

🅲 hạng C
28
More Agents, More Samples, or More Thinking? Untangling Coordination, Ensembling, and Test-Time Compute in Multi-Agent LLMs
📄 Paper arXiv cs.AI 13 giờ trước

arXiv:2611.17610v1 Announce Type: new Abstract: Evaluations of multi-agent language models often compare communicating architectures against a single greedy pass. This practice confounds coordination with sample multiplicity and inference compute. We synthesize nine primary empi…

🅲 hạng C
28
On Private Knowledge Retrieval and Private Inference for secure multi-agent systems
📄 Paper arXiv cs.CR 5 ngày trước

arXiv:2611.13696v1 Announce Type: new Abstract: Multi-agent systems provide a convenient framework for the massive deployment of agents based on Large Language Models (LLMs). It provides a sound framework where agents can interact with existing software to perform actions in an …

🅲 hạng C
28
Tail-risk-aware Agent Safety: Constraining Tail Risks Across Multi-step LLM Agent Workflows
📄 Paper arXiv cs.AI 6 ngày trước

arXiv:2611.10411v1 Announce Type: new Abstract: LLM-based agents extend LLMs with tool calls, memory, planning, and retrieval to act in external environments, but they are vulnerable to attacks at different workflow steps. Existing defenses, including prompt engineering and post…

🅲 hạng C
28
Tracing MCP Security Vulnerabilities: Constructing and Analyzing an MCP Security Vulnerability Dataset
📄 Paper arXiv cs.CR 6 ngày trước

arXiv:2611.11104v1 Announce Type: new Abstract: As a core protocol for integrating large language models (LLMs) with external data sources and tools, the Model Context Protocol (MCP) has quickly become a key interface in emerging AI agent systems. As the MCP ecosystem expands, r…

🅲 hạng C
27
OverServe: Resource Amplification Attacks on SaaS Agents via Tool--Prompt Co-Design
🛡️ Tin bảo mật arXiv cs.CR 3 ngày trước

OverServe: kỹ thuật resource amplification attack nhắm vào SaaS agents — tenant xấu dùng prompt khai thác quyết định 'khi nào agent tự dừng' để kéo dài execution và đốt tài nguyên model/tool của provider (đặc biệt model subscription-based). Đáng chú ý với ai xây/thẩm định AI agent trong SaaS vì dễ bị lạm dụng chi phí.

🅲 hạng C
⚪ Mức độ: CHƯA ĐỦ DỮ LIỆU · RCE
Dùng để khai thác: chạy mã tuỳ ý trên máy chủ — thường là chiếm quyền cao nhất, không cần người dùng bấm gì
Hậu quả: chiếm toàn bộ máy chủ
🎯 Kẻ tấn công làm gì (từng bước)
  1. Tìm đúng đường dẫn/API có xử lý dữ liệu người dùng gửi lên
  2. Gửi payload đã chế để máy chủ thực thi lệnh/mã
  3. Dùng quyền của tiến trình web để đọc file, cài backdoor, mở kênh ra ngoài
🔍 Dấu hiệu bị tấn công (để kiểm tra log)
  • Tiến trình web (php-fpm/nginx/java) sinh tiến trình con bất thường: sh, bash, curl, wget
  • Kết nối ra ngoài từ máy chủ tới IP lạ đúng lúc có request lạ
  • File mới xuất hiện trong thư mục tải lên/tạm
🛡 Phòng ngay (làm được hôm nay)
  • Vá ngay bản có lỗi (ưu tiên số 1)
  • Chặn tạm đường dẫn/API bị lỗi ở WAF/reverse proxy nếu chưa vá được
  • Chạy tiến trình web bằng user quyền thấp, tắt các hàm shell (disable_functions) nếu dùng PHP
  • Bật giám sát tiến trình con của web server để phát hiện lệnh lạ
→ Theo dõi và vá theo lịch; chưa thấy dấu hiệu bị khai thác.
27
Policy and Disclosure: 2026 Edition
💥 Lỗ hổng / Exploit Project Zero 3 ngày trước thảo luận

Since 2021, Project Zero has followed a “90 + 30” vulnerability disclosure policy, allowing vendors up to 90 days to remediate a vulnerability plus an extra 30 days “soak time” for patch rollout. Our goal was to encourage faster patching and give vendors time to deliver patches t…

🅰 hạng A
26
LLMs are Vulnerable to Malicious Requests with Indistinguishable Intent
📄 Paper arXiv cs.CR 13 giờ 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
26
AI Reviews Are Biased Towards the Appendix
📄 Paper HF Daily Papers 3 ngày trước

Large language models (LLMs) are increasingly used by reviewers to highlight weaknesses during the reviewing process, and their outputs often shape the reviews that humans submit. Prior work studies how LLM reviews align with human ones, yet we know little about which parts of a …

🅲 hạng C
25
G-CAST: Governed Continual Self-Training for E-Commerce Agents under Evolving Demands
📄 Paper arXiv cs.AI 13 giờ trước cũng có trên: HF Daily Papers

arXiv:2611.17814v1 Announce Type: new Abstract: With the rapid advancement of Large Language Models (LLMs), an increasing number of users are turning to LLM-powered agents for conversational shopping in e-commerce. However, the user needs continue to change as new product catego…

🅲 hạng C
24
Actor, Judge, or Teacher? Understanding Decision Models in Long-Horizon Agents
📄 Paper HF Daily Papers 4 ngày trước

Decision models answer a question with a typed output, such as a choice among options with a probability for each, so one model can choose an agent's actions, evaluate actions already taken, or score every action to teach a policy with reinforcement learning. Prior work has studi…

🅲 hạng C
24
Advancing Quantization for R&R Models with Microscaling Formats
📄 Paper arXiv cs.AI 13 giờ trước

arXiv:2611.17501v1 Announce Type: new Abstract: Recommendation and Ranking (R&R) models, particularly in digital advertising, require massive inference throughput but exhibit extreme sensitivity to precision loss. Unlike Large Language Models (LLMs) that frequently tolerate quan…

🅲 hạng C
24
Is Abliteration Getting Harder? Not Uniformly: Refusal Removal by Direction-Based Ablation and ARA in Heretic on the Qwen3.8, Gemma 4, Olmo 3.1 and Mistral Small 3.2 LLMs Versus Their Predecessors
📄 Paper arXiv cs.AI 13 giờ trước

arXiv:2611.17846v1 Announce Type: new Abstract: Abliteration, a fine-tuning-free weight edit, removes refusal from open-weight large language models (LLMs). The developer of Arbitrary-Rank Ablation (ARA), an optimisation-based variant, reports that direction-based abliteration, …

🅲 hạng C
23
molly554/whiteboard-ai-tutor
🛠️ Công cụ AI GitHub mới (#llm) 4 ngày trước ⭐ 84 HTML 📄 MIT

🧩 Chức năng: Công cụ quanh mô hình ngôn ngữ (prompt/agent)

Open-source AI tutor that teaches on a whiteboard. Benchmark of 6 LLMs (Claude, Gemini, DeepSeek): 288 lessons drawn stroke by stroke, blind-reviewed. 白板 AI 导师 / AI 老师讲课评测

🅲 hạng C
23
Privacy Leakage from Recurrent State in LLM Inference
📄 Paper arXiv cs.CR 3 ngày trước

arXiv:2611.16048v1 Announce Type: new Abstract: Personalized LLM services place private facts and application instructions into a prompt, then retain representations of that context for later inference. Latent-state LLMs compress part of this history into fixed-size recurrent st…

🅲 hạng C
23
Browse tools.simonwillison.net by tag
📰 Tin AI Simon Willison 4 ngày trước

Browse tools.simonwillison.net by tag I've been wanting a better way to browse my collection of vibe-coded tools (244 and counting) for a while. I had a few Fable 5.2 credits left over before my weekly reset so I set it to work using tree-sitter to analyze the code for JavaScript…

🅲 hạng C
23
Anthropic's IPO roadshow video
🎬 AI Video Simon Willison 5 ngày trước

Anthropic's IPO roadshow video Anthropic kicked off their IPO roadshow yesterday, and their roadshow video has some previously unreleased information about how their business works. The section on "Approach to compute" with co-founder Tom Brown is particularly interesting, descri…

🅲 hạng C
23
Ask HN: What are the most useful, concrete AI skills to independently learn?
🤖 AI Agent / MCP Hacker News 2 ngày trước ▲ 87 · 💬 51

Specifically, what are the most useful, concrete skills to learn as an individual to take economic advantage of the advances in LLMs, agents and associated new technological capabilities? What is accessible to learn just from the internet and experimentation, without a degree pro…

🅲 hạng C