↓Skip to main content

🏢 MBZUAI

Self-Guiding Exploration for Combinatorial Problems
·2441 words·12 mins· loading · loading
Natural Language Processing Large Language Models 🏢 MBZUAI
LLMs excel at reasoning tasks, but their application to combinatorial problems (CPs) is underexplored. This paper introduces Self-Guiding Exploration (SGE), a novel prompting strategy that significan…
Remove that Square Root: A New Efficient Scale-Invariant Version of AdaGrad
·1874 words·9 mins· loading · loading
AI Generated Machine Learning Optimization 🏢 MBZUAI
KATE: A new scale-invariant AdaGrad variant achieves state-of-the-art convergence without square roots, outperforming AdaGrad and matching/exceeding Adam’s performance.
LFME: A Simple Framework for Learning from Multiple Experts in Domain Generalization
·2799 words·14 mins· loading · loading
Machine Learning Domain Generalization 🏢 MBZUAI
LFME: A novel framework improves domain generalization by training multiple expert models alongside a target model, using logit regularization for enhanced performance.
DLAD: Improving Logits-based Detector without Logits from Black-box LLMs
·2559 words·13 mins· loading · loading
Natural Language Processing Large Language Models 🏢 MBZUAI
DALD: A novel framework for black-box LLM text detection, achieving state-of-the-art performance without relying on source model logits, by aligning surrogate model distributions.