会员服务
PDR: A Plug-and-Play Positional Decay Framework for LLM Pre-training Data Detection
Arxiv
0+阅读 · 1月11日
Forgetting Similar Samples: Can Machine Unlearning Do it Better?
Arxiv
0+阅读 · 1月11日
Learning Design-Score Manifold to Guide Diffusion Models for Offline Optimization
Arxiv
0+阅读 · 1月12日
Dr. Zero: Self-Evolving Search Agents without Training Data
Arxiv
0+阅读 · 1月11日
What Can We Actually Steer? A Multi-Behavior Study of Activation Control
Arxiv
0+阅读 · 1月11日
An information-matching approach to optimal experimental design and active learning
Arxiv
0+阅读 · 1月12日
Metaphors are a Source of Cross-Domain Misalignment of Large Reasoning Models
Metaphors are a Source of Cross-Domain Misalignment of Large Reasoning Models
Arxiv
0+阅读 · 1月12日
GlobalRAG: Enhancing Global Reasoning in Multi-hop Question Answering via Reinforcement Learning
Arxiv
0+阅读 · 1月12日
Zero-Day Audio DeepFake Detection via Retrieval Augmentation and Profile Matching
Arxiv
0+阅读 · 1月9日
Compact Example-Based Explanations for Language Models
Arxiv
0+阅读 · 1月7日
How Training Data Shapes the Use of Parametric and In-Context Knowledge in Language Models
Arxiv
0+阅读 · 1月7日
Low Resource Reconstruction Attacks Through Benign Prompts
Arxiv
0+阅读 · 1月7日
Zero-Day Audio DeepFake Detection via Retrieval Augmentation and Profile Matching
Arxiv
0+阅读 · 1月7日
Difficulty Controlled Diffusion Model for Synthesizing Effective Training Data
Arxiv
0+阅读 · 1月7日
VERSE: Visual Embedding Reduction and Space Exploration. Clustering-Guided Insights for Training Data Enhancement in Visually-Rich Document Understanding
Arxiv
0+阅读 · 1月8日
参考链接
微信扫码咨询专知VIP会员
Top