cs.LG（2025-01-05）

📊 共 7 篇论文 | 🔗 2 篇有代码

🎯 兴趣领域导航

#	题目	一句话要点	标签	🔗
1	Representation Learning of Lab Values via Masked AutoEncoders	提出Lab-MAE，利用掩码自编码器进行电子病历中缺失实验室值的表征学习与补全。	representation learning masked autoencoder MAE
2	DPO Kernels: A Semantically-Aware, Kernel-Enhanced, and Divergence-Rich Paradigm for Direct Preference Optimization	DPO-Kernels：一种语义感知、核增强、多样性丰富的直接偏好优化范式	DPO direct preference optimization large language model
3	LeetDecoding: A PyTorch Library for Exponentially Decaying Causal Linear Attention with CUDA Implementations	LeetDecoding：基于PyTorch的CUDA加速指数衰减因果线性注意力库	linear attention large language model	✅
4	Representation Convergence: Mutual Distillation is Secretly a Form of Regularization	互蒸馏作为正则化手段，提升强化学习策略对无关特征的鲁棒性	reinforcement learning distillation

#	题目	一句话要点	标签	🔗
5	Efficient Deployment of Large Language Models on Resource-constrained Devices	FedSpine：面向资源受限设备的LLM高效联邦部署框架	large language model
6	HALO: Hadamard-Assisted Lower-Precision Optimization for LLMs	HALO：面向LLM的Hadamard辅助低精度优化，实现高效量化微调	large language model	✅
7	Transformers Simulate MLE for Sequence Generation in Bayesian Networks	Transformer通过模拟MLE，在贝叶斯网络中实现序列生成	large language model