cập nhật 5 phút trước

⏱ Độ trễ thật của tab này: trung bình 6.3 giờ sau khi CVE được công bố (n = 5287, sớm nhất 0.1 giờ). Số âm nghĩa là ta thấy trước ngày công bố.

28
VAST: Compiling Grounded Vulnerability Scenarios into Verifiable, Multi-Step Environments for Evaluating LLM Red-Team Agents
📄 Paper arXiv cs.CR 5 ngày trước

arXiv:2611.19465v1 Announce Type: new Abstract: Evaluating Large Language Model (LLM) security agents requires reproducible environments whose workflow difficulty can be varied and whose failures can be diagnosed. We present VAST (Verifiable Adversarial Scenario Training & Evalu…

🅲 hạng C
27
CompoundingKernel: Self-Improving from Zero to First
📄 Paper arXiv cs.AI 5 ngày trước cũng có trên: HF Daily Papers

arXiv:2611.19230v1 Announce Type: new Abstract: We present CompoundingKernel, a self-improving agent system that writes GPU kernels by search. As of 2026-11-11, its generated kernels rank #1 on the public L1, L2 and Quantization boards of SOL-ExecBench (NVIDIA B200). They have h…

🅲 hạng C
27
The Resolution Hypothesis: Agent Evolution's Key Is Identifying Improvements
📄 Paper HF Daily Papers 5 ngày trước

Agent evolution follows a simple loop: a proposer writes candidate artifacts (prompts, skills, memories) and selection accepts those that score higher. This loop has shown promise, yet practitioners face a puzzle: methods trade wins across papers, and evolution often fails to boo…

🅲 hạng C
27
QDA: Closed-Loop LLM Quantization Design with Agents
📄 Paper HF Daily Papers 5 ngày trước

Post-training quantization is the standard way to serve large language models at low cost. However, the best quantization recipe differs across models and precisions. As a result, experts must redesign it for every new model, which is time-consuming and often sub-optimal. Fortuna…

🅲 hạng C
25
Toward a Science of AI Agent Societies
📄 Paper arXiv cs.AI 2 ngày trước

arXiv:2611.25404v1 Announce Type: new Abstract: AI agents are rapidly evolving from isolated personal assistants into networked actors that interact with one another at scale. We envision the emergence of AI agent societies, with their own social and economic dynamics, as a new …

🅲 hạng C
25
GRAFT: Execution-Guided Genetic Recombination for Agent Design
📄 Paper arXiv cs.AI 3 ngày trước

arXiv:2611.23509v1 Announce Type: new Abstract: Automated agent design searches for executable architectures that improve LLM-based agent performance, with evolutionary methods using language-model crossover to recombine promising mechanisms for task decomposition, model coordin…

🅲 hạng C
24
TeleChat to Xing4: Advancing Frontier Agentic Models on Domestic Chips and Frameworks
📄 Paper HF Daily Papers 5 ngày trước cũng có trên: arXiv cs.AI

We present Xing 4, a series of large language models that extends the TeleChat lineage through coordinated advances in model architecture and training infrastructure on Ascend. This report describes the design principles and systems engineering behind the series, spanning archite…

🅲 hạng C
24
Can an LLM Fix a Language Model Without Training It?
📄 Paper arXiv cs.AI 2 ngày trước

arXiv:2611.25613v1 Announce Type: new Abstract: Although an external language model can select and test repairs to another model's weights, behavior can also be recovered through fixed numerical edits. We assessed the value of interactive selection in a controlled study of Trans…

🅲 hạng C
23
GUARD: Stabilizing Recursive Self-Improvement with Regret Guidance
📄 Paper arXiv cs.AI 3 ngày trước

arXiv:2611.23476v1 Announce Type: new Abstract: Autonomous AI agents pair a frozen base model with an editable harness of prompts, skills, tool rules, and memory. Recursive self-improvement (RSI) adapts this harness from experience without weight updates, yet self-editing withou…

🅲 hạng C
23
From Sorry to QED: Refining LLM Math Proofs via Angelic Execution
📄 Paper arXiv cs.AI 4 ngày trước

arXiv:2611.22103v1 Announce Type: new Abstract: Large language models (LLMs) can now generate Lean proof scripts that capture the high-level structure of many mathematical arguments. However, these scripts frequently fail on complex theorems: a single missing side condition or m…

🅲 hạng C
23
From Token Changes to Task Failures: Bit Flips in a Shared KV-Cache Block
📄 Paper arXiv cs.CR 4 ngày trước

arXiv:2611.22243v1 Announce Type: new Abstract: In LLM serving systems with prefix caching, each KV block of a shared system prompt is held in GPU memory as a single physical copy, which every request with that prompt reads. Our earlier work measured change and showed that one b…

🅲 hạng C
23
What Moves a LoCoMo Score? Memory Context, Answer Prompts and Judges
📄 Paper arXiv cs.AI 5 ngày trước

arXiv:2611.19199v1 Announce Type: new Abstract: LoCoMo is a widely reported benchmark for long-term conversational memory, but published scores mix memory quality with choices of answer prompt, judge and baseline layout that are rarely reported together. We measure how much thes…

🅲 hạng C
23
A First Look at System One Decision Models for Malware Detection
📄 Paper arXiv cs.CR 5 ngày trước

arXiv:2611.20699v1 Announce Type: new Abstract: Malware detection needs scores calibrated at a fixed false positive rate (FPR) and decisions an analyst can read: feature-based detectors give the first, large language models (LLMs) on decompiled code the second, through generated…

🅲 hạng C
22
A Verifier from the Past: Test-Time Compute for Time Series Foundation Models
📄 Paper arXiv cs.AI 2 ngày trước

arXiv:2611.25171v1 Announce Type: new Abstract: Test-time compute (TTC) improves language models by spending additional inference compute to generate, verify, and select among candidate outputs, but this recipe has not transferred directly to time series forecasting. A language …

🅲 hạng C
22
LoongForge-Embodied: A Unified Framework for Training Heterogeneous Embodied Models
📄 Paper arXiv cs.AI 2 ngày trước

arXiv:2611.25460v1 Announce Type: new Abstract: Training Vision-Language-Action (VLA) models and World-Action Models (WAMs) is computationally expensive, while their heterogeneous architectures and fragmented training stacks make system optimizations difficult to reuse across mo…

🅲 hạng C
22
AeroBelief: Budgeted First-Action Verification for Evidence-Guided Aerial Sensing
📄 Paper arXiv cs.AI 2 ngày trước

arXiv:2611.25568v1 Announce Type: new Abstract: Aerial sensing must combine uncertain overhead context with current observations while budgeting prediction, movement and image processing. AeroBelief uses the next executable action as the unit of verification against reference sc…

🅲 hạng C
22
GPU-Accelerated Secure Inference for Drug--Target Interaction Prediction
📄 Paper arXiv cs.CR 2 ngày trước

arXiv:2611.25664v1 Announce Type: new Abstract: Deep learning models that predict drug--target interactions (DTI) can speed up drug discovery, but the candidate drugs they score are valuable intellectual property that their owners cannot reveal. Secure multiparty computation (MP…

🅲 hạng C
21
Training Vision Models for LLM-Driven Weight-Space Programming
📄 Paper HF Daily Papers 3 ngày trước

It has been demonstrated that, in certain settings, large language models can perform weight-space programming to directly modify the weights of neural networks. However, it is unknown how dedicated training of these networks can contribute to the successes of weight-space progra…

🅲 hạng C
21
Asymmetric Optical Image Watermarking
📄 Paper arXiv cs.CR 3 ngày trước

arXiv:2611.23350v1 Announce Type: new Abstract: Watermark detectors held by a provider cannot be checked by third parties, and a party that can detect a mark can usually also forge it. We study a two-layer image watermark in which the Fourier transform at the core of the pipelin…

🅲 hạng C
21
Smoothing and Learning for Movement-Robust RSS-Based Spoofing Detection
📄 Paper arXiv cs.CR 3 ngày trước

arXiv:2611.23854v1 Announce Type: new Abstract: Physical-layer spoofing detection identifies an attacker who impersonates a legitimate wireless user by checking whether transmissions claiming the same identity originate from different locations, using the received signal strengt…

🅲 hạng C
21
The Limit Is in the Input, Not the Model: State-Dependent Command Safety in Natural-Language Interfaces to Industrial Controllers
📄 Paper arXiv cs.CR 3 ngày 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