cập nhật 15 phút trước

⏱ Độ trễ thật của tab này: trung bình 7.2 giờ sau khi CVE được công bố (n = 5387, sớm nhất 0.1 giờ). Số âm nghĩa là ta thấy trước ngày công bố.

27
Tree-Ising: Repairing Language Models with Bit-Level Ising Edits
📄 Paper HF Daily Papers 6 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
AI Reviews Are Biased Towards the Appendix
📄 Paper HF Daily Papers 3 ngày trước

Large language models (LLMs) are increasingly used by reviewers to highlight weaknesses during the reviewing process, and their outputs often shape the reviews that humans submit. Prior work studies how LLM reviews align with human ones, yet we know little about which parts of a …

🅲 hạng C
25
Intent-Aware Protection for AI Agents
📄 Paper arXiv cs.CR 3 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
LoRAWeaver: Directional Adapter Routing in LoRA Representation Space
📄 Paper arXiv cs.AI 11 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
OCTAVE: A Capability-Resolved Benchmark for End-to-End HTML5 Game Generation
📄 Paper arXiv cs.AI 11 giờ trước

arXiv:2611.17681v1 Announce Type: new Abstract: Complete game generation requires agents to coordinate rules, interactions, states, and presentation over a long development process. A single aggregate score obscures which capabilities are implemented and which requirements remai…

🅲 hạng C
24
TReC: Evaluating Move-Induced State Changes in Instruction-Conditioned Go
📄 Paper arXiv cs.AI 11 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
Representational Geometry as a Diagnostic for Adversarial Robustness
📄 Paper arXiv cs.CR 11 giờ trước

arXiv:2611.17684v1 Announce Type: cross Abstract: Adversarial training is a common technique for making neural networks more robust to small, intentionally crafted input perturbations, and robustness is commonly measured through accuracy under attack. By exploring inter-network …

🅲 hạng C
23
Privacy Leakage from Recurrent State in LLM Inference
📄 Paper arXiv cs.CR 3 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
21
UniSpark: Exploring Hybrid Attention for Multimodal Understanding and Generation
📄 Paper HF Daily Papers 4 ngày trước

We present UniSpark-9B, a unified multimodal model that investigates hybrid attention for visual understanding and generation. Built on a Mixture-of-Transformer-Experts backbone with approximately 9B parameters in each branch, UniSpark combines full attention for explicit token i…

🅲 hạng C
20
Building World Models from Intrinsic Motivation
📄 Paper arXiv cs.AI 5 ngày trước

arXiv:2611.13113v1 Announce Type: new Abstract: In the absence of extrinsic rewards or task supervision, human behavior is guided by intrinsic motivation---neural signals that drive curiosity, creativity, play, and other essential facets of intelligence. Intrinsic motivation pla…

🅲 hạng C
20
TERM: Target-Free Jailbreak Defense via Entropy-Guided Residual Modulation
📄 Paper arXiv cs.CR 5 ngày trước

arXiv:2611.13577v1 Announce Type: new Abstract: Large language models (LLMs) have shown strong capabilities across diverse applications, making both safety and utility essential for reliable deployment. Jailbreak attacks use diverse forms of adversarial prompting to bypass align…

🅲 hạng C
20
POEM: Output-Relative Block-Sparse Attention for Long-Context Prefilling
📄 Paper arXiv cs.CR 5 ngày trước

arXiv:2611.13578v1 Announce Type: new Abstract: Large language models (LLMs) have shown strong capabilities across diverse applications, making both safety and utility essential for reliable deployment. Jailbreak attacks use diverse forms of adversarial prompting to bypass align…

🅲 hạng C
20
DexForesight: Grounding Dexterous Policies with Implicit Spatial Foresight
📄 Paper HF Daily Papers 6 ngày trước

Dexterous manipulation requires rich spatial foresight over both the surrounding scene and the robot hand. Existing world-action models often acquire such predictive knowledge through explicit future prediction. We introduce DexForesight, a framework that learns spatial foresight…

🅲 hạng C
20
Emergent Biological Capabilities in a Foundation Model for Molecular Interactions
📄 Paper HF Daily Papers 6 ngày trước

Foundation models trained on large unlabeled corpora develop emergent capabilities, tasks that go beyond the training objective. We extend this paradigm to multimodal biology with Bonbon, a foundation model for molecular interactions trained on approximately 3.2 trillion tokens o…

🅲 hạng C