cs.LG（2024-12-31）

📊 共 15 篇论文 | 🔗 3 篇有代码

🎯 兴趣领域导航

支柱九：具身大模型 (Embodied Foundation Models) (7) 支柱二：RL算法与架构 (RL & Architecture) (6 🔗2) 支柱八：物理动画 (Physics-based Animation) (1) 支柱三：空间感知与语义 (Perception & Semantics) (1 🔗1)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (7 篇)

#题目一句话要点标签🔗⭐
1 Low-Rank Adaptation for Foundation Models: A Comprehensive Review LoRA综述：全面回顾低秩适应方法在通用基础模型上的应用与发展 large language model foundation model
2 Towards Sustainable Large Language Model Serving 从碳排放角度研究LLM服务，为可持续大语言模型服务铺平道路 large language model
3 Differentiable Prompt Learning for Vision Language Models 提出可微Prompt学习(DPL)方法，自动优化视觉语言模型中的Prompt配置。 large language model
4 Finding Missed Code Size Optimizations in Compilers using LLMs 利用LLM辅助的差分测试发现编译器中遗漏的代码大小优化 large language model
5 Prune 'n Predict: Optimizing LLM Decision-making with Conformal Prediction 提出CROQ与CP-OPT以优化LLM决策过程 large language model
6 Generalizing Trust: Weak-to-Strong Trustworthiness in Language Models 研究大型语言模型中弱到强泛化，探索可信属性的迁移能力 large language model
7 Towards Pattern-aware Data Augmentation for Temporal Knowledge Graph Completion 提出Booster，一种模式感知的数据增强方法，用于提升时序知识图谱补全任务性能。 TAMP

🔬 支柱二：RL算法与架构 (RL & Architecture) (6 篇)

#题目一句话要点标签🔗⭐
8 Toward Information Theoretic Active Inverse Reinforcement Learning 提出信息论主动逆强化学习框架，提升人机交互效率 reinforcement learning inverse reinforcement learning
9 Understanding and Mitigating Bottlenecks of State Space Models through the Lens of Recency and Over-smoothing 通过分析SSM的近因偏见和过平滑问题，提出极化技术以提升长程依赖建模能力。 SSM state space model ✅
10 Towards Unraveling and Improving Generalization in World Models 通过随机微分方程分析和改进世界模型的泛化能力 reinforcement learning world model
11 KAE: Kolmogorov-Arnold Auto-Encoder for Representation Learning 提出Kolmogorov-Arnold自编码器(KAE)，提升表征学习在检索、分类和去噪任务中的性能。 representation learning ✅
12 Beyond Introspection: Reinforcing Thinking via Externalist Behavioral Feedback 提出DRR框架，通过外部行为反馈增强LLM的推理能力，克服自省幻觉。 distillation large language model
13 Goal Recognition using Actor-Critic Optimization DRACO：利用Actor-Critic优化进行目标识别，无需人工设计和离散表示。 reinforcement learning deep reinforcement learning

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

#题目一句话要点标签🔗⭐
14 diffIRM: A Diffusion-Augmented Invariant Risk Minimization Framework for Spatiotemporal Prediction over Graphs 提出diffIRM框架以解决图结构时空预测中的OOD问题 spatiotemporal

🔬 支柱三：空间感知与语义 (Perception & Semantics) (1 篇)

#题目一句话要点标签🔗⭐
15 Outlier-Robust Training of Machine Learning Models 提出自适应交替算法，用于机器学习模型在离群点下的鲁棒训练 scene reconstruction ✅

⬅️ 返回 cs.LG 首页 · 🏠 返回主页