cs.LG（2025-09-05）

📊 共 23 篇论文

🎯 兴趣领域导航

支柱二：RL算法与架构 (RL & Architecture) (12) 支柱九：具身大模型 (Embodied Foundation Models) (9) 支柱八：物理动画 (Physics-based Animation) (1) 支柱一：机器人控制 (Robot Control) (1)

🔬 支柱二：RL算法与架构 (RL & Architecture) (12 篇)

#题目一句话要点标签🔗⭐
1 Greener Deep Reinforcement Learning: Analysis of Energy and Carbon Efficiency Across Atari Benchmarks 评估深度强化学习算法在Atari游戏中能源消耗与碳排放效率，为绿色AI提供基准。 reinforcement learning deep reinforcement learning DRL
2 Deep Reinforcement Learning for Ranking Utility Tuning in the Ad Recommender System at Pinterest 提出DRL-PUT框架，利用深度强化学习优化Pinterest广告推荐系统中排序效用函数。 reinforcement learning deep reinforcement learning DRL
3 FinXplore: An Adaptive Deep Reinforcement Learning Framework for Balancing and Discovering Investment Opportunities FinXplore：一种自适应深度强化学习框架，用于平衡和发现投资机会 reinforcement learning deep reinforcement learning DRL
4 Beyond I-Con: Exploring New Dimension of Distance Measures in Representation Learning Beyond I-Con：探索表征学习中距离度量的新维度，提升聚类与降维效果 representation learning contrastive learning
5 Self-Aligned Reward: Towards Effective and Efficient Reasoners 提出自对齐奖励(SAR)，提升LLM推理精度与效率，降低计算成本。 reinforcement learning PPO large language model
6 An Arbitration Control for an Ensemble of Diversified DQN variants in Continual Reinforcement Learning 提出ACED-DQN，通过仲裁控制多样化DQN集成解决持续强化学习中的灾难性遗忘问题 reinforcement learning deep reinforcement learning
7 MambaLite-Micro: Memory-Optimized Mamba Inference on MCUs MambaLite-Micro：面向MCU的内存优化Mamba模型推理引擎 Mamba
8 PLanTS: Periodicity-aware Latent-state Representation Learning for Multivariate Time Series PLanTS：提出周期感知的潜在状态表征学习框架，用于多元时间序列分析。 representation learning
9 SpikingBrain: Spiking Brain-inspired Large Models SpikingBrain：受脑启发的线性注意力大模型，提升长文本处理效率。 linear attention large language model
10 Shift Before You Learn: Enabling Low-Rank Representations in Reinforcement Learning 提出基于转移后继测度的低秩强化学习方法，提升目标条件RL性能 reinforcement learning
11 Pre-Forgettable Models: Prompt Learning as a Native Mechanism for Unlearning 提出Pre-Forgettable模型，通过Prompt学习实现模型原生可遗忘性，解决数据隐私合规问题。 distillation foundation model
12 Topology-Aware Graph Reinforcement Learning for Dynamic Routing in Cloud Networks 提出拓扑感知图强化学习，解决云网络动态路由优化问题 reinforcement learning

🔬 支柱九：具身大模型 (Embodied Foundation Models) (9 篇)

#题目一句话要点标签🔗⭐
13 Multimodal Foundation Model-Driven User Interest Modeling and Behavior Analysis on Short Video Platforms 提出基于多模态基础模型的用户兴趣建模方法，用于短视频平台行为分析与推荐。 foundation model multimodal
14 DreamPRM-1.5: Unlocking the Potential of Each Instance for Multimodal Process Reward Model Training DreamPRM-1.5：通过实例重加权提升多模态过程奖励模型的训练效果 multimodal
15 Probabilistic operator learning: generative modeling and uncertainty quantification for foundation models of differential equations 提出GenICON，通过生成建模和不确定性量化提升微分方程基础模型的泛化能力。 foundation model
16 ModalSurv: Investigating opportunities and limitations of multimodal deep survival learning in prostate and bladder cancer ModalSurv：探索多模态深度生存学习在前列腺癌和膀胱癌中的机遇与局限 multimodal
17 veScale: Consistent and Efficient Tensor Programming with Eager-Mode SPMD veScale：通过Eager模式SPMD实现一致且高效的张量编程 large language model
18 Neural Breadcrumbs: Membership Inference Attacks on LLMs Through Hidden State and Attention Pattern Analysis 提出memTrace，通过分析LLM内部表征进行成员推断攻击，揭示潜在隐私泄露。 large language model
19 On Using Large-Batches in Federated Learning 探索联邦学习中大批量训练的优势与挑战，提升模型泛化能力 multimodal
20 KVCompose: Efficient Structured KV Cache Compression with Composite Tokens KVCompose：基于组合Token的高效结构化KV缓存压缩方法 large language model
21 Revolution or Hype? Seeking the Limits of Large Models in Hardware Design 探索硬件设计中大模型的局限性：一场革命还是炒作？ large language model

🔬 支柱八：物理动画 (Physics-based Animation) (1 篇)

#题目一句话要点标签🔗⭐
22 Deep Learning-Enhanced for Amine Emission Monitoring and Performance Analysis in Industrial Carbon Capture Plants 利用深度学习预测胺排放和性能，优化工业碳捕集 AMP

🔬 支柱一：机器人控制 (Robot Control) (1 篇)

#题目一句话要点标签🔗⭐
23 On the Learnability of Distribution Classes with Adaptive Adversaries 研究自适应对抗下的分布类可学习性，揭示其与传统对抗学习的差异 manipulation

⬅️ 返回 cs.LG 首页 · 🏠 返回主页