cs.LG（2025-03-14）

📊 共 30 篇论文 | 🔗 2 篇有代码

🎯 兴趣领域导航

支柱九：具身大模型 (Embodied Foundation Models) (14 🔗1) 支柱二：RL算法与架构 (RL & Architecture) (11 🔗1) 支柱八：物理动画 (Physics-based Animation) (3) 支柱七：动作重定向 (Motion Retargeting) (1) 支柱一：机器人控制 (Robot Control) (1)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (14 篇)

#题目一句话要点标签🔗⭐
1 Empowering Time Series Analysis with Synthetic Data: A Survey and Outlook in the Era of Foundation Models 综述：合成数据赋能时序分析，展望基础模型时代的应用 large language model foundation model
2 LLMPerf: GPU Performance Modeling meets Large Language Models LLMPerf：利用大语言模型进行GPU性能建模，提升程序成本分析效率 large language model
3 How Can Time Series Analysis Benefit From Multiple Modalities? A Survey and Outlook 综述多模态时间序列分析（MM4TSA），探索如何利用其他模态提升时间序列分析性能。 foundation model multimodal
4 D3: Diversity, Difficulty, and Dependability-Aware Data Selection for Sample-Efficient LLM Instruction Tuning 提出D3方法以解决大规模数据集中的样本选择问题 large language model instruction following
5 FedALT: Federated Fine-Tuning through Adaptive Local Training with Rest-of-World LoRA 提出FedALT，通过自适应局部训练和RoW LoRA实现个性化联邦微调 large language model
6 Test-Time Training Provably Improves Transformers as In-context Learners 提出基于梯度测试时训练方法，提升Transformer上下文学习能力并降低样本复杂度 foundation model
7 Performance Analysis of Decentralized Federated Learning Deployments 分析去中心化联邦学习部署性能，揭示网络拓扑和数据分布的影响 large language model
8 Understanding the Trade-offs in Accuracy and Uncertainty Quantification: Architecture and Inference Choices in Bayesian Neural Networks 贝叶斯神经网络中精度与不确定性量化的权衡研究：架构与推断选择的影响 multimodal
9 Reasoning-Grounded Natural Language Explanations for Language Models 提出一种基于推理过程的大语言模型自然语言可解释性技术 large language model
10 PrivacyScalpel: Enhancing LLM Privacy via Interpretable Feature Intervention with Sparse Autoencoders PrivacyScalpel：利用可解释特征干预和稀疏自编码器增强LLM隐私 large language model
11 Don't Forget It! Conditional Sparse Autoencoder Clamping Works for Unlearning 基于条件稀疏自编码器钳制的LLM知识遗忘技术 large language model
12 A Survey of Cross-domain Graph Learning: Progress and Future Directions 综述跨域图学习进展与未来方向，旨在实现真正的图基础模型。 foundation model ✅
13 Generative Modeling for Mathematical Discovery 提出基于LLM驱动的遗传算法FunSearch，用于辅助数学家进行数学发现。 large language model
14 From Dionysius Emerges Apollo -- Learning Patterns and Abstractions from Perceptual Sequences 提出基于组块和抽象的序列学习模型，用于从感知序列中发现模式和层次结构。 large language model

🔬 支柱二：RL算法与架构 (RL & Architecture) (11 篇)

#题目一句话要点标签🔗⭐
15 Contextual Similarity Distillation: Ensemble Uncertainties with a Single Model 提出上下文相似性蒸馏，用单模型高效估计深度集成的不确定性，提升强化学习探索效率。 reinforcement learning offline reinforcement learning distillation
16 Unicorn: A Universal and Collaborative Reinforcement Learning Approach Towards Generalizable Network-Wide Traffic Signal Control Unicorn：一种通用协作强化学习方法，用于可泛化的全网络交通信号控制 reinforcement learning contrastive learning
17 SPECTra: Scalable Multi-Agent Reinforcement Learning with Permutation-Free Networks SPECTra：基于无排列网络的可扩展多智能体强化学习 reinforcement learning curriculum learning ✅
18 Crash Severity Analysis of Child Bicyclists using Arm-Net and MambaNet 利用ARM-Net和MambaNet分析儿童自行车事故严重程度，MambaNet表现更优。 predictive model Mamba
19 A Review of DeepSeek Models' Key Innovative Techniques DeepSeek模型创新技术综述：低成本实现媲美顶尖闭源LLM的性能 reinforcement learning large language model
20 Creating a Good Teacher for Knowledge Distillation in Acoustic Scene Classification 声学场景分类中，研究教师模型属性对知识蒸馏学生模型性能的影响 distillation
21 OPTIMUS: Predicting Multivariate Outcomes in Alzheimer's Disease Using Multi-modal Data amidst Missing Values OPTIMUS：利用多模态数据和可解释AI预测阿尔茨海默病中的多变量结果 predictive model multimodal
22 Technologies on Effectiveness and Efficiency: A Survey of State Spaces Models 综述性论文：深入剖析状态空间模型（SSM）的有效性和效率技术 Mamba SSM state space model
23 Enabling Weak Client Participation via On-device Knowledge Distillation in Heterogeneous Federated Learning 提出基于设备端知识蒸馏的异构联邦学习方法，解决弱客户端参与问题。 distillation
24 Statistical Impossibility and Possibility of Aligning LLMs with Human Preferences: From Condorcet Paradox to Nash Equilibrium 揭示LLM对齐人类偏好的统计极限：从孔多塞悖论到纳什均衡 reinforcement learning large language model
25 Residual Policy Gradient: A Reward View of KL-regularized Objective 提出残差策略梯度(RPG)，扩展残差Q学习到策略梯度方法，用于策略定制。 reinforcement learning imitation learning

🔬 支柱八：物理动画 (Physics-based Animation) (3 篇)

#题目一句话要点标签🔗⭐
26 Integrating Dynamical Systems Modeling with Spatiotemporal scRNA-seq Data Analysis 综述：整合动力系统建模与时空单细胞RNA测序数据分析 spatiotemporal
27 Brain Effective Connectivity Estimation via Fourier Spatiotemporal Attention 提出基于傅里叶时空注意力的FSTA-EC方法，用于提升脑功能磁共振成像有效连接估计的准确性。 spatiotemporal
28 Federated Koopman-Reservoir Learning for Large-Scale Multivariate Time-Series Anomaly Detection 提出FedKO：一种基于联邦学习的Koopman-Reservoir模型，用于大规模多元时间序列异常检测。 spatiotemporal

🔬 支柱七：动作重定向 (Motion Retargeting) (1 篇)

#题目一句话要点标签🔗⭐
29 CoLLMLight: Cooperative Large Language Model Agents for Network-Wide Traffic Signal Control CoLLMLight：用于网络级交通信号控制的协同大语言模型智能体 spatial relationship spatiotemporal large language model

🔬 支柱一：机器人控制 (Robot Control) (1 篇)

#题目一句话要点标签🔗⭐
30 Low-cost Real-world Implementation of the Swing-up Pendulum for Deep Reinforcement Learning Experiments 提出一种低成本倒立摆实验平台，用于弥合深度强化学习中的Sim-to-Real差距 sim-to-real reinforcement learning deep reinforcement learning

⬅️ 返回 cs.LG 首页 · 🏠 返回主页