cập nhật 1 giờ trước

⏱ Độ trễ thật của tab này: trung bình 6.8 giờ sau khi CVE được công bố (n = 5426, sớm nhất 0.1 giờ). Số âm nghĩa là ta thấy trước ngày công bố.

29
Mem2Fair: Online Fairness Control for Text-to-Video Generation via Fair Memory
📄 Paper arXiv cs.AI 2 ngày trước

arXiv:2611.19215v1 Announce Type: new Abstract: Text-to-video (T2V) generation models can produce systematically unfair demographic outcomes when prompts leave protected attributes unspecified. For example, prompts describing nurses without specifying gender may disproportionate…

🅲 hạng C
28
The Resolution Hypothesis: Agent Evolution's Key Is Identifying Improvements
📄 Paper HF Daily Papers 2 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
28
Security of Deep Research Agents: Attack Surface, Defenses, Benchmarks, and Agenda Seeding against Research Planning
📄 Paper arXiv cs.CR 2 ngày trước

arXiv:2611.20045v1 Announce Type: new Abstract: Deep research and agentic search agents autonomously retrieve content from the open web and generate reports with citations, and their outputs are often taken directly as research conclusions. This paper systematizes security resea…

🅲 hạng C
27
What LLM Agents See and What Users Say
📄 Paper arXiv cs.AI 3 ngày trước

arXiv:2611.17639v1 Announce Type: new Abstract: Before an LLM agent cancels a reservation or deletes a folder for a user, what stops it from acting on a false assumption: what its tools show, or what the user says? In this work, we study this gap with presupposition theory from …

🅲 hạng C
26
GUARD: Stabilizing Recursive Self-Improvement with Regret Guidance
📄 Paper arXiv cs.AI 13 giờ 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
26
PolicySniff: Inferring Agent Policy States through Shared-GPU Execution
📄 Paper arXiv cs.CR 6 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
Auditable Authorization for LLM Agents: Empirical Evidence Across Model, Harness, and Workflow Layers
📄 Paper arXiv cs.AI 2 ngày trước

arXiv:2611.19237v1 Announce Type: new Abstract: Human approval provides effective control only if rejected work stops and authorized work completes. We formalize refusal safety and bounded authorization completeness, and audit three execution boundaries: the model selects a call…

🅲 hạng C
25
SPLICE: Structured Prompt Local Iterative Combinatorial Evolution
📄 Paper arXiv cs.AI 2 ngày trước

arXiv:2611.18883v1 Announce Type: new Abstract: Natural language optimization (NLO) improves an agent by having a model critique and rewrite its natural language policy parameters: system prompts, tool instructions, reasoning directives, few-shot demonstrations. These artifacts …

🅲 hạng C
25
Threat Landscape of Vibe Coding and the Road to Secure Autonomy
📄 Paper arXiv cs.CR 1 ngày trước

arXiv:2611.21822v1 Announce Type: new Abstract: Vibe coding is reshaping software development, enabling faster implementation and allowing individuals with limited programming experience to describe requirements in natural language while AI agents write, test, and deploy the cod…

🅲 hạng C
25
From Token Changes to Task Failures: Bit Flips in a Shared KV-Cache Block
📄 Paper arXiv cs.CR 1 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
25
PIPETTE: Prompt Injection Purple-Teaming for Robust LLM Agents
📄 Paper HF Daily Papers 1 ngày trước cũng có trên: arXiv cs.CR

Large language model (LLM) agents routinely interact with untrusted external data, exposing them to indirect prompt injection (IPI) attacks that hijack agents into executing attacker-chosen tasks. To uncover these vulnerabilities at scale, automated red-teaming methods often leve…

🅲 hạng C
24
Jeveration: System One Models Now Generate
📄 Paper HF Daily Papers 1 ngày trước

System One models such as Jev return decisions through fast, parallel prediction. In this paper, we introduce Jeveration, showing that the same decision paradigm can also generate high-quality images in a single forward pass. Jeveration builds on a frozen vision-language model, r…

🅲 hạng C
24
Can you make models believe false facts in the prompt?
📄 Paper arXiv cs.AI 13 giờ trước

arXiv:2611.23938v1 Announce Type: new Abstract: Making LLMs believe false facts is useful for alignment evaluations and for building model organisms. Previous work explored doing so with Synthetic Document Finetuning (SDF). But SDF requires fine-tuning one new model for each set…

🅲 hạng C
24
Intent-Aware Protection for AI Agents
📄 Paper arXiv cs.CR 6 ngày trước

arXiv:2611.16122v1 Announce Type: new Abstract: AI agents increasingly access enterprise data, invoke powerful tools, and act with limited human oversight. Existing security defenses primarily detect prompt injection, harmful outputs, or suspicious behavior, but do not determine…

🅲 hạng C
24
What Moves a LoCoMo Score? Memory Context, Answer Prompts and Judges
📄 Paper arXiv cs.AI 2 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
Privacy Leakage from Recurrent State in LLM Inference
📄 Paper arXiv cs.CR 6 ngày trước

arXiv:2611.16048v1 Announce Type: new Abstract: Personalized LLM services place private facts and application instructions into a prompt, then retain representations of that context for later inference. Latent-state LLMs compress part of this history into fixed-size recurrent st…

🅲 hạng C
22
Language Models Can Rewrite Their Own Weights
📄 Paper arXiv cs.AI 1 ngày trước

arXiv:2611.21885v1 Announce Type: new Abstract: Language models are trained once, frozen and deployed: improving one again takes another training run, and a prompt, a scaffold or extra computation at inference is gone once the task ends. A frontier model has been shown to edit t…

🅲 hạng C
22
Fine-Grained Japanese Clinical Named Entity Recognition: A Controlled Comparison of BERT and Open-Weight Large Language Models
📄 Paper arXiv cs.AI 1 ngày trước

arXiv:2611.22219v1 Announce Type: new Abstract: Background: Generative large language models (LLMs) offer an alternative to encoder-based clinical named entity recognition (NER), but their performance on fine-grained Japanese extraction tasks and sensitivity to generation settin…

🅲 hạng C
21
When the Spec Sheet Lies: Injection Attacks on LLM-Generated Compliance Artifacts
📄 Paper arXiv cs.CR 2 ngày trước

arXiv:2611.19090v1 Announce Type: new Abstract: As Large Language Models (LLMs) become used more commonly across a variety of fields, a growing body of research has highlighted particular concerns surrounding model security. Notably, prompt injection attacks, where malicious ins…

🅲 hạng C
21
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 3 ngày 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
21
Beyond Answering: Learning to Speak First and Act in Streaming Video
📄 Paper HF Daily Papers 4 ngày trước

Most AI assistants respond only when prompted, remaining silent even when events in the video stream offer meaningful opportunities for conversation. To move beyond purely user-driven interaction, we present AURA 2.0, a streaming video large language model that introduces three c…

🅲 hạng C