cập nhật 6 phút trước

⏱ Độ trễ thật của tab này: trung bình 6.8 giờ sau khi CVE được công bố (n = 5457, sớm nhất 0.1 giờ). Số âm nghĩa là ta thấy trước ngày công bố.

27
Memory Tax: Dissecting the Net Value of Agent Memory
📄 Paper HF Daily Papers 24 ngày trước

Memory is increasingly used in long-horizon agents, yet the net value of building memory over direct retrieval from raw history remains largely unclear. Through a large-scale evaluation across multiple benchmarks and model backbones, we find many dedicated memory systems underper…

🅲 hạng C
27
Advances in Agentic Computer Graphics
📄 Paper HF Daily Papers 17 ngày trước

Computer graphics is built on code. Large language model (LLM) agents can now generate graphics code and iteratively refine it through various feedback. This capability opens the door to a potentially new paradigm of computer graphics, with graphics algorithms increasingly built …

🅲 hạng C
27
Tree-Ising: Repairing Language Models with Bit-Level Ising Edits
📄 Paper HF Daily Papers 8 ngày trước

Weight-space programming, introduced in recent MIT work by Takehi et al., enables a frontier LLM agent to read a network's weights and write a repair. Their GPT-6 Astra agent corrected the reasoning loops of Qwen3.5-0.8B without gradients, using 158 agent turns, 89.7 million toke…

🅲 hạng C
26
What Does Multi-Turn Reinforcement Learning Really Improve in LLM Agents?
📄 Paper HF Daily Papers 13 ngày trước

While multi-turn agentic reinforcement learning (RL) excels on long-horizon tasks, its performance gains remain highly sensitive to interaction constraints. Evaluating across sampling breadth K and interaction depth T via Pass@(K,T), we show that trajectory-level RL advantages de…

🅲 hạng C
26
PCBRouteAgent: How Close Are LLM Agents to Becoming PCB Routing Engineers?
📄 Paper HF Daily Papers 12 ngày trước

Automating printed circuit board (PCB) routing requires coordinating board-level connectivity decisions with geometric execution under strict design rules. Although recent agents can interact with native routing tools, reliable routing on complex boards remains challenging. We pr…

🅲 hạng C
26
Retrieval Augmented Generation Compositionality
📄 Paper HF Daily Papers 28 ngày trước

Retrieval Augmented Generation (RAG) grounds large language models in external knowledge. Modern pipelines combine embedding, chunking, retrieval, reranking, query augmentation, context compression, and agentic verification, yet these techniques are typically evaluated one at a t…

🅲 hạng C
26
Evolving Agent Interfaces
📄 Paper HF Daily Papers 29 ngày trước

Computer-use agents are limited by the interfaces humans give them. An agent may type a spreadsheet cell by cell when a single line of code would do, or print all two thousand lines of a file through a shell to find one function. Both the interfaces and the knowledge of when to u…

🅲 hạng C
26
Beyond Watching and Searching: Benchmarking Agents for Video-centered Deep Research Reports
📄 Paper HF Daily Papers 27 ngày trước

Deep research agents synthesize open-web evidence into long reports, but current benchmarks do not evaluate research centered on a source video. We introduce Video-Centered Deep Research Report Generation. Given a video and user request, an agent identifies research questions, re…

🅲 hạng C
26
RIDAR: Safeguarding Research Integrity in Autonomous Research Agents via Automated Detection and Runtime Intervention
📄 Paper HF Daily Papers 27 ngày trước

Autonomous research agents are increasingly used to automate large portions of scientific research, yet they are prone to research integrity violations (RIVs): behaviors that compromise the validity, reproducibility, or faithfulness of the research process. We present RIDAR, a fr…

🅲 hạng C
26
RSI Arena: Towards Recursive Self-Improvement with Human Feedback
📄 Paper HF Daily Papers 25 ngày trước

Can AI agents autonomously train useful language models and improve them using feedback from users? We introduce RSI Arena, a two-stage study of autonomous post-training as a step toward recursive self-improvement. Ten frontier models act as training agents, each starting from NV…

🅲 hạng C
26
ParetoSaddler: Pareto-Guided Optimization of Agent Harnesses for Quality and Cost
📄 Paper HF Daily Papers 17 ngày trước

Harness optimization can substantially improve LLM agents without modifying model weights, but existing methods primarily optimize response quality. Extending them to joint quality--cost optimization introduces a challenge: applying the deployment preference during intermediate u…

🅲 hạng C
26
Mobile AI Factory: Building Specialized Language Models for Your Mobile Device
📄 Paper HF Daily Papers 8 ngày trước

Small language models (SLMs) running on mobile devices can keep user data private and work without connectivity. But they perform well only when fine-tuned for the target task and sized for the target device, a process that today requires repeated expert tuning. We present {\it M…

🅲 hạng C
26
"Shall We Approve?" Consequence-Aware Approval for Autonomous Coding Agents
📄 Paper arXiv cs.CR 15 ngày trước cũng có trên: HF Daily Papers

arXiv:2610.40753v1 Announce Type: new Abstract: Coding agents increasingly rely on automated approvers to decide whether a proposed shell command may run. Existing approvers, from rule-based checkers to guardrail models and deployed permission classifiers, base their decisions o…

🅲 hạng C
26
SwarmEvolve: Latent Agent Swarms for Evolutionary Discovery and Optimization
📄 Paper HF Daily Papers 13 ngày trước

LLM-guided evolutionary search uses AI agents to improve programs for mathematical and computational problems by iteratively refining solutions and combining high-scoring methods. However, the process is often slow because the LLM generates a long textual chain of thought for eac…

🅲 hạng C
26
Authorization Drift in LLM Agents: Evaluating Boundary Compliance in Multi-Step Tool Use
📄 Paper arXiv cs.CR 8 ngày trước cũng có trên: HF Daily Papers

arXiv:2611.09113v1 Announce Type: new Abstract: LLM agents can exceed their authorization boundaries while pursuing user goals without an adversary. We formalize this phenomenon as authorization drift across data, tool, scope, and time, and introduce the Authorization Drift Benc…

🅲 hạng C
26
When Expert Disagreement Hurts: Prestige and Selectivity in LLM Decision Revision
📄 Paper HF Daily Papers 6 ngày trước

When language models revise a decision after another agent disagrees, neither reversal counts nor final accuracy shows whether they are fixing mistakes or abandoning correct answers. We introduce a controlled two-pass audit that holds the question, the evidence, and the model's i…

🅲 hạng C
26
Thinking-in-Bets at Test Time Reduces Sycophantic Revisions in LLM Agents
📄 Paper HF Daily Papers 23 ngày trước

Language model agents often abandon correct work when a user challenges it, yet simply telling them to stand firm makes them ignore real mistakes. Once an agent reads a complaint, it tends to discount its own earlier checks in order to agree with the user. We introduce Thinking-i…

🅲 hạng C
26
Tools Need Not Speak in Text: Latent Context Compression for Agents
📄 Paper HF Daily Papers 14 ngày trước

Large language model (LLM) agents typically append raw tool outputs to their context. Over multi-step interactions, long and noisy observations incur substantial context overhead and dilute the agent's attention to task-relevant information. Existing token pruning and summarizati…

🅲 hạng C
26
CPU-ARC: Agentic Reasoning Caching for on-CPU Reasoning Reuse
📄 Paper HF Daily Papers 12 ngày trước

Large language models reinvent the wheel all the time. They conflate task-solving with the execution of the solutions, which forces them to solve every problem from scratch. We show that once a task has been reasoned through, the application of its solution to a new problem insta…

🅲 hạng C
26
How Much Feedback Is Needed to Break Template Protection? The Geometric Vulnerability of Biometric Authentication
📄 Paper arXiv cs.CR 29 ngày trước

arXiv:2610.25118v1 Announce Type: new Abstract: Biometric authentication systems increasingly use privacy-preserving mechanisms to protect sensitive templates. However, the authentication interface itself may still leak information that enables reconstruction attacks. Prior atta…

🅲 hạng C
26
AdaFedXplain: Adaptive Federated Learning with Explainable AI for Ransomware Detection
📄 Paper arXiv cs.CR 29 ngày trước

arXiv:2610.25318v1 Announce Type: new Abstract: Attacks using ransomware have developed into complex multiphase campaigns that target vital infrastructure, resulting in annual damages of billions of dollars. Conventional detection techniques are ineffective against zero-day vari…

🅲 hạng C
👁 Trúng danh sách theo dõi: kw:zero-day