cs.LG（2025-07-13）

📊 共 14 篇论文

🎯 兴趣领域导航

支柱九：具身大模型 (Embodied Foundation Models) (6) 支柱二：RL算法与架构 (RL & Architecture) (4) 支柱八：物理动画 (Physics-based Animation) (2) 支柱七：动作重定向 (Motion Retargeting) (2)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (6 篇)

#题目一句话要点标签🔗⭐
1 Continental-scale habitat distribution modelling with multimodal earth observation foundation models 利用多模态地球观测基础模型进行洲际尺度生境分布建模 foundation model multimodal
2 LLMs Meet Cross-Modal Time Series Analytics: Overview and Directions 综述LLM在跨模态时间序列分析中的应用，并探讨未来发展方向 large language model
3 Compute Requirements for Algorithmic Innovation in Frontier AI Models 研究前沿AI模型算法创新所需的计算资源，揭示算力限制对创新的影响 large language model
4 Sub-Scaling Laws: On the Role of Data Density and Training Strategies in LLMs 研究数据密度与训练策略对大语言模型次尺度定律的影响 large language model
5 Cultivating Pluralism In Algorithmic Monoculture: The Community Alignment Dataset 提出Community Alignment数据集，解决LLM在多元文化偏好对齐上的挑战。 large language model
6 Latent Sensor Fusion: Multimedia Learning of Physiological Signals for Resource-Constrained Devices 提出基于隐空间融合的多模态生理信号学习方法，适用于资源受限设备。 multimodal

🔬 支柱二：RL算法与架构 (RL & Architecture) (4 篇)

#题目一句话要点标签🔗⭐
7 Fine-tuning Large Language Model for Automated Algorithm Design 通过微调大语言模型，实现自动化算法设计的性能提升与泛化 direct preference optimization large language model
8 Meta-Reinforcement Learning for Fast and Data-Efficient Spectrum Allocation in Dynamic Wireless Networks 提出基于元强化学习的频谱动态分配方法，提升5G/6G网络资源利用率。 reinforcement learning deep reinforcement learning DRL
9 Efficient Molecular Conformer Generation with SO(3)-Averaged Flow Matching and Reflow 提出SO(3)平均Flow Matching与Reflow，加速分子构象生成。 flow matching distillation
10 Scalpel vs. Hammer: GRPO Amplifies Existing Capabilities, SFT Replaces Them 对比强化学习与监督微调：揭示LLM推理能力训练的差异与影响 reinforcement learning large language model

🔬 支柱八：物理动画 (Physics-based Animation) (2 篇)

#题目一句话要点标签🔗⭐
11 MB-RIRs: a Synthetic Room Impulse Response Dataset with Frequency-Dependent Absorption Coefficients 提出基于多频带吸收系数的合成房间脉冲响应数据集MB-RIRs，提升单声道语音增强效果。 PULSE
12 Machine-Precision Prediction of Low-Dimensional Chaotic Systems 利用高精度计算，实现对低维混沌系统动力学的机器精度预测 spatiotemporal

🔬 支柱七：动作重定向 (Motion Retargeting) (2 篇)

#题目一句话要点标签🔗⭐
13 Universal Physics Simulation: A Foundational Diffusion Approach 提出一种通用物理模拟的扩散模型，无需先验方程即可从边界条件数据中学习物理定律。 spatial relationship
14 Federated Learning with Graph-Based Aggregation for Traffic Forecasting 提出基于图聚合的联邦学习方法，用于解决交通预测中空间依赖问题。 spatial relationship

⬅️ 返回 cs.LG 首页 · 🏠 返回主页