cập nhật 6 phút trước

⏱ Độ trễ thật của tab này: trung bình 6.8 giờ sau khi CVE được công bố (n = 5467, sớm nhất 0.1 giờ). Số âm nghĩa là ta thấy trước ngày công bố.

24
When Retrieval Is Not Enough: Obsolete Facts in Agent Memory
📄 Paper arXiv cs.AI 22 ngày trước

arXiv:2610.36308v1 Announce Type: new Abstract: Memory-agent benchmarks often collapse retrieval and answer generation into one score, making it hard to tell whether a failure comes from missing an update or from not using an update that was already retrieved. We examine this di…

🅲 hạng C
24
Pricing Cross-Bank Context in Federated Anti-Money Laundering
📄 Paper arXiv cs.CR 22 ngày trước

arXiv:2610.35996v1 Announce Type: new Abstract: Most graph-based Anti-Money Laundering (AML) methods assume access to a pooled transaction graph containing all accounts and payments, which in reality does not exist. Federated learning offers the necessary privacy guarantees but …

🅲 hạng C
24
WebAnything: From the Wild Web to Scalable Agent Environments
📄 Paper arXiv cs.AI 20 ngày trước

arXiv:2610.37786v1 Announce Type: new Abstract: Web agents require diverse, executable, and resettable website environments for reproducible training and evaluation. Human-assisted mirrors demand per-site engineering, while generative construction can diverge from specified real…

🅲 hạng C
24
Seq@K: Evaluating Language Models Under Sequential Feedback
📄 Paper arXiv cs.AI 20 ngày trước

arXiv:2610.37859v1 Announce Type: new Abstract: Language-model agents differ fundamentally from prior intelligent systems in their ability to learn from natural language in context. This capability enables agents to iteratively refine their answers in response to feedback, a pre…

🅲 hạng C
24
VerifiedLab: Harnessing Knowns for Unknowns
📄 Paper arXiv cs.AI 20 ngày trước

arXiv:2610.37923v1 Announce Type: new Abstract: Across human history, knowledge has undergone several transitions, from orality to writing and then to digital archives. At each transition, humans established new conventions to preserve knowledge for the new era. Nowadays, as res…

🅲 hạng C
24
Poisoned Skills: Propagating Jailbreaks Through Agent Markdown Files
📄 Paper arXiv cs.CR 20 ngày trước

arXiv:2610.37738v1 Announce Type: new Abstract: Modern agent frameworks allow users to provide Markdown files containing system prompts, tool descriptions, and skill definitions (e.g., SKILL.md), which are incorporated into the agent's context during initialization. In this work…

🅲 hạng C
24
Appreciation Teaches Unauthorized Execution: Self-Evolution Degrades the Robustness of Personal Agents to Indirect Prompt Injection
📄 Paper arXiv cs.CR 20 ngày trước

arXiv:2610.39122v1 Announce Type: new Abstract: Self-evolving personal agents act for users and learn from user feedback. However, if an agent obeys an instruction that an untrusted third party hides in content, the third party gains control of the agent. Such indirect prompt in…

🅲 hạng C
24
RoboArchitect: Automating the Design of Robotic Agents with Persistent Memory
📄 Paper HF Daily Papers 20 ngày trước

Recent work shows that a frozen robot policy performs far better inside an agent harness, where control code decides when to query the policy, consult vision tools, or command the arm directly. Yet today these harnesses are built by hand and redesigned for each new robot or bench…

🅲 hạng C
24
Towards Portable Specification of the Agent Harness
📄 Paper arXiv cs.AI 17 ngày trước

arXiv:2610.40917v1 Announce Type: new Abstract: Large language models act only through the infrastructure that surrounds them. That infrastructure, the harness, determines what a model can accomplish and how it fails, yet it has no settled definition. To address this gap, we dec…

🅲 hạng C
24
The First Few Tokens Matter: Understanding Chain-of-Thought Controllability
📄 Paper arXiv cs.AI 17 ngày trước

arXiv:2610.40939v1 Announce Type: new Abstract: Chain-of-thought (CoT) monitoring is a promising way to oversee increasingly capable agents. It succeeds in part because current models appear limited in their ability to control what they verbalize in their reasoning traces. We in…

🅲 hạng C
24
P-ContextBench: From Personal Fact Recall to Personalized Agent Behavior
📄 Paper arXiv cs.AI 17 ngày trước

arXiv:2610.40959v1 Announce Type: new Abstract: Large language models (LLMs) are increasingly used for personal tasks, yet evaluating their safe use of personal context remains difficult because such contexts are private, fragmented, dynamic, and socially situated. We introduce …

🅲 hạng C
24
RTL-Interact: Training EDA-Interactive Models for Multi-Task RTL Design
📄 Paper arXiv cs.AI 17 ngày trước

arXiv:2610.41225v1 Announce Type: new Abstract: Large language models offer a promising route to register-transfer-level (RTL) design, but reliable implementations require checking and revision based on electronic design automation (EDA) tools' feedback. Existing RTL-specialized…

🅲 hạng C
24
Towards Elastic Least Privilege for Agent Planning
📄 Paper arXiv cs.CR 17 ngày trước

arXiv:2610.40981v1 Announce Type: new Abstract: General-purpose AI agents require authority that changes between tasks and during execution. Common deployments trade broad permissions that expose unnecessary authority and narrow permissions that raise approval requests. Existing…

🅲 hạng C
24
Can Coding Agents Sustain Continual Evolution without Forgetting?
📄 Paper HF Daily Papers 17 ngày trước

Recently, harness self-evolution has emerged as an important paradigm for improving agent performance. Prior studies show that evolving the agent harness can effectively enhance agent capabilities without modifying the foundation model's parameters. We find that existing approach…

🅲 hạng C
24
Learning to bid and bundle: a posterior-sampling agent for online retail
📄 Paper arXiv cs.AI 16 ngày trước

arXiv:2611.00165v1 Announce Type: new Abstract: An online retailer launching a perishable product bids for sponsored-search clicks, discounts a bundle with an always-stocked staple and displays its remaining stock, which raises willingness to pay. It knows neither how clicks res…

🅲 hạng C
24
AuroraWorld: Structured 4D World Modeling for Exo-to-Ego Video Generation in Robot Manipulation
📄 Paper arXiv cs.AI 16 ngày trước

arXiv:2611.00177v1 Announce Type: new Abstract: Egocentric observations provide a robot-centric view of manipulation processes, capturing contact interactions and object-state changes that are important for visuomotor policy learning but difficult to observe from external camera…

🅲 hạng C
24
MemU-Bench: Benchmarking GUI Agents on Long-Horizon User Interactions
📄 Paper arXiv cs.AI 16 ngày trước

arXiv:2611.00191v1 Announce Type: new Abstract: Existing Graphical User Interface (GUI) benchmarks predominantly rely on precise and stateless user instructions. In everyday usage, however, instructions are often \emph{vague and history-dependent}: users omit the exact details o…

🅲 hạng C