cs.LG（2025-09-12）

📊 共 23 篇论文 | 🔗 1 篇有代码

🎯 兴趣领域导航

支柱九：具身大模型 (Embodied Foundation Models) (9 🔗1) 支柱二：RL算法与架构 (RL & Architecture) (9) 支柱八：物理动画 (Physics-based Animation) (2) 支柱六：视频提取与匹配 (Video Extraction) (1) 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) (9 篇)

#题目一句话要点标签🔗⭐
1 Test-Time Warmup for Multimodal Large Language Models 提出测试时预热方法，提升多模态大语言模型在复杂推理任务上的性能。 large language model multimodal
2 DyKen-Hyena: Dynamic Kernel Generation via Cross-Modal Attention for Multimodal Intent Recognition DyKen-Hyena：通过跨模态注意力动态生成卷积核，用于多模态意图识别 multimodal
3 LoFT: Parameter-Efficient Fine-Tuning for Long-tailed Semi-Supervised Learning in Open-World Scenarios 提出LoFT框架，通过高效参数微调解决开放世界长尾半监督学习问题 foundation model
4 Characterizing the Efficiency of Distributed Training: A Power, Performance, and Thermal Perspective 深入剖析LLM分布式训练效率：从功耗、性能和热管理的角度进行全面评估 large language model ✅
5 SME-TEAM: Leveraging Trust and Ethics for Secure and Responsible Use of AI and LLMs in SMEs 提出SME-TEAM框架，解决中小企业安全合规地使用AI和LLM的问题 large language model
6 RFSeek and Ye Shall Find RFSeek：利用LLM自动提取RFC协议逻辑的可视化摘要，提升协议理解 large language model
7 Neural Scaling Laws for Deep Regression 研究深度回归模型神经标度律，揭示数据量与模型容量对性能的影响 large language model
8 Limited Reference, Reliable Generation: A Two-Component Framework for Tabular Data Generation in Low-Data Regimes ReFine：低数据量下表格数据生成双组件框架，提升生成质量与下游任务性能 large language model
9 SciML Agents: Write the Solver, Not the Solution SciML Agents：利用LLM生成求解器代码，而非直接预测科学计算结果 instruction following

🔬 支柱二：RL算法与架构 (RL & Architecture) (9 篇)

#题目一句话要点标签🔗⭐
10 Data distribution impacts the performance and generalisability of contrastive learning-based foundation models of electrocardiograms 提出CAPE心电图基础模型，解决对比学习在心电图数据分布偏移下的泛化性问题。 contrastive learning foundation model
11 Inpainting-Guided Policy Optimization for Diffusion Large Language Models 提出IGPO：利用Inpainting引导扩散LLM的强化学习策略优化 reinforcement learning large language model
12 Generalizing Beyond Suboptimality: Offline Reinforcement Learning Learns Effective Scheduling through Random Data 提出CDQAC算法，通过离线强化学习从随机数据中学习高效作业调度策略 reinforcement learning offline RL offline reinforcement learning
13 BenchECG and xECG: a benchmark and baseline for ECG foundation models BenchECG：心电图（ECG）基础模型的标准化评测基准与xECG基线模型 representation learning foundation model
14 CrunchLLM: Multitask LLMs for Structured Business Reasoning and Outcome Prediction CrunchLLM：用于结构化商业推理和结果预测的多任务LLM predictive model large language model foundation model
15 Coordinated Reinforcement Learning Prefetching Architecture for Multicore Systems 提出一种协同强化学习预取架构CRL-Pythia，解决多核系统中的冗余预取问题。 reinforcement learning
16 Federated Multi-Agent Reinforcement Learning for Privacy-Preserving and Energy-Aware Resource Management in 6G Edge Networks 提出Fed-MARL框架，解决6G边缘网络中隐私保护和节能的资源管理问题 reinforcement learning
17 Prototypical Contrastive Learning For Improved Few-Shot Audio Classification 提出原型对比学习框架，提升小样本音频分类性能 contrastive learning
18 Why and How Auxiliary Tasks Improve JEPA Representations 通过辅助任务改进JEPA表征，解决表征坍塌问题并提升表征质量 latent dynamics model-based RL representation learning

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

#题目一句话要点标签🔗⭐
19 GraphCSVAE: Graph Categorical Structured Variational Autoencoder for Spatiotemporal Auditing of Physical Vulnerability Towards Sustainable Post-Disaster Risk Reduction 提出GraphCSVAE模型，用于时空审计灾后物理脆弱性，助力可持续风险降低。 spatiotemporal
20 Symbolic Feedforward Networks for Probabilistic Finite Automata: Exact Simulation and Learnability 提出基于符号前馈网络的概率有限自动机精确模拟与学习方法 differentiable simulation

🔬 支柱六：视频提取与匹配 (Video Extraction) (1 篇)

#题目一句话要点标签🔗⭐
21 Using LLMs for Late Multimodal Sensor Fusion for Activity Recognition 利用LLM进行活动识别的后期多模态传感器融合 Ego4D large language model multimodal

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

#题目一句话要点标签🔗⭐
22 Physics-informed sensor coverage through structure preserving machine learning 提出基于结构保持机器学习的物理信息传感器覆盖方法，用于自适应源定位。 structure preservation

🔬 支柱一：机器人控制 (Robot Control) (1 篇)

#题目一句话要点标签🔗⭐
23 Prompt Injection Attacks on LLM Generated Reviews of Scientific Publications 揭示LLM生成论文评审中的提示注入攻击风险，并评估其有效性 manipulation

⬅️ 返回 cs.LG 首页 · 🏠 返回主页