cs.LG（2024-02-10）

📊 共 10 篇论文 | 🔗 2 篇有代码

🎯 兴趣领域导航

#	题目	一句话要点	标签	🔗
1	OpenFedLLM: Training Large Language Models on Decentralized Private Data via Federated Learning	提出OpenFedLLM以解决去中心化私有数据训练问题	large language model instruction following	✅
2	Generating Chain-of-Thoughts with a Pairwise-Comparison Approach to Searching for the Most Promising Intermediate Thought	提出基于成对比较的方法以优化链式思维生成	large language model chain-of-thought
3	$L^*LM$: Learning Automata from Examples using Natural Language Oracles	提出$L^*LM$算法以提高DFAs学习的样本效率	large language model
4	A Tale of Tails: Model Collapse as a Change of Scaling Laws	提出模型崩溃理论框架以应对合成数据的影响	large language model
5	Fiddler: CPU-GPU Orchestration for Fast Inference of Mixture-of-Experts Models	提出Fiddler以解决Mixture-of-Experts模型的资源限制问题	large language model	✅
6	Whispers in the Machine: Confidentiality in Agentic Systems	提出LLM代理系统的机密性评估方法以应对安全威胁	large language model

#	题目	一句话要点	标签
7	Principled Penalty-based Methods for Bilevel Reinforcement Learning and RLHF	提出基于惩罚的方法解决双层强化学习及人类反馈问题	reinforcement learning inverse reinforcement learning RLHF
8	Solving Deep Reinforcement Learning Tasks with Evolution Strategies and Linear Policy Networks	通过进化策略和线性策略网络解决深度强化学习任务	reinforcement learning deep reinforcement learning
9	In-Context Data Distillation with TabPFN	提出上下文数据蒸馏方法以解决TabPFN的数据规模限制问题	distillation foundation model
10	Informativeness of Reward Functions in Reinforcement Learning	提出奖励信息量标准以加速强化学习收敛	reinforcement learning reward design