cs.LG（2025-09-04）

📊 共 22 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) (12 🔗2) 支柱二：RL算法与架构 (RL & Architecture) (8) 支柱一：机器人控制 (Robot Control) (2)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (12 篇)

#题目一句话要点标签🔗⭐
1 Multimodal Deep Learning for ATCO Command Lifecycle Modeling and Workload Prediction 提出多模态深度学习框架，用于空管指挥生命周期建模与工作负荷预测 multimodal
2 Delta Activations: A Representation for Finetuned Large Language Models 提出Delta Activations，通过激活值变化表征微调LLM，实现模型聚类、选择与合并。 large language model ✅
3 IPA: An Information-Reconstructive Input Projection Framework for Efficient Foundation Model Adaptation 提出IPA：一种信息重构的输入投影框架，用于高效地适应基础模型 foundation model ✅
4 PagedEviction: Structured Block-wise KV Cache Pruning for Efficient Large Language Model Inference PagedEviction：用于高效大语言模型推理的结构化块级KV缓存剪枝 large language model
5 COBRA: Multimodal Sensing Deep Learning Framework for Remote Chronic Obesity Management via Wrist-Worn Activity Monitoring COBRA：基于腕部活动监测的多模态深度学习框架，用于远程慢性肥胖管理 multimodal
6 MEUV: Achieving Fine-Grained Capability Activation in Large Language Models via Mutually Exclusive Unlock Vectors MEUV：通过互斥解锁向量实现大语言模型中的细粒度能力激活 large language model
7 ChronoGraph: A Real-World Graph-Based Multivariate Time Series Dataset ChronoGraph：一个基于真实微服务系统的图结构多元时间序列数据集 foundation model
8 Characteristic Energy Behavior Profiling of Non-Residential Buildings 提出一种数据驱动的非住宅建筑能耗行为建模方法，用于评估能源系统中断的影响。 multimodal
9 One-Embedding-Fits-All: Efficient Zero-Shot Time Series Forecasting by a Model Zoo ZooCast：通过模型动物园实现高效的零样本时间序列预测 foundation model
10 KubeGuard: LLM-Assisted Kubernetes Hardening via Configuration Files and Runtime Logs Analysis KubeGuard：利用LLM分析配置与日志，增强Kubernetes安全性 large language model
11 Privacy Risks in Time Series Forecasting: User- and Record-Level Membership Inference 针对时间序列预测模型的用户和记录级别成员推理隐私风险研究 large language model
12 TAGAL: Tabular Data Generation using Agentic LLM Methods TAGAL：利用Agentic LLM方法生成表格数据，提升下游机器学习任务性能。 large language model

🔬 支柱二：RL算法与架构 (RL & Architecture) (8 篇)

#题目一句话要点标签🔗⭐
13 Towards a Unified View of Large Language Model Post-Training 统一大语言模型后训练视角，提出混合后训练算法HPT，提升数学推理能力。 reinforcement learning large language model
14 RL's Razor: Why Online Reinforcement Learning Forgets Less 揭示RL微调优于SFT的原因：在线强化学习具备更少的遗忘性 reinforcement learning large language model foundation model
15 Rethinking the long-range dependency in Mamba/SSM and transformer models 从理论角度分析Mamba/SSM和Transformer模型中的长程依赖建模能力 Mamba SSM
16 Wavelet Fourier Diffuser: Frequency-Aware Diffusion Model for Reinforcement Learning 提出Wavelet Fourier Diffuser，解决离线强化学习中轨迹频率偏移问题。 reinforcement learning offline reinforcement learning
17 Data-Augmented Quantization-Aware Knowledge Distillation 提出数据增强感知的量化知识蒸馏方法，提升低比特模型精度 distillation
18 Connections between reinforcement learning with feedback,test-time scaling, and diffusion guidance: An anthology 揭示强化学习、测试时缩放与扩散引导的内在联系，提出重采样对齐方法。 reinforcement learning
19 Resource-Aware Neural Network Pruning Using Graph-based Reinforcement Learning 提出基于图强化学习的资源感知神经网络剪枝方法，提升剪枝效率和性能。 reinforcement learning
20 Parking Availability Prediction via Fusing Multi-Source Data with A Self-Supervised Learning Enhanced Spatio-Temporal Inverted Transformer 提出SST-iTransformer，融合多源数据和自监督学习，提升停车位可用性预测精度。 representation learning MAE

🔬 支柱一：机器人控制 (Robot Control) (2 篇)

#题目一句话要点标签🔗⭐
21 Unobtrusive In-Situ Measurement of Behavior Change by Deep Metric Similarity Learning of Motion Patterns 提出基于深度度量相似性学习的非侵入式行为变化测量方法，用于XR环境中用户行为分析。 manipulation affordance
22 DRtool: An Interactive Tool for Analyzing High-Dimensional Clusterings DRtool：用于分析高维聚类结果的交互式工具，诊断降维伪结构。 manipulation

⬅️ 返回 cs.LG 首页 · 🏠 返回主页