cs.LG（2024-11-05）

📊 共 22 篇论文 | 🔗 7 篇有代码

🎯 兴趣领域导航

支柱九：具身大模型 (Embodied Foundation Models) (13 🔗5) 支柱二：RL算法与架构 (RL & Architecture) (6 🔗2) 支柱五：交互与反应 (Interaction & Reaction) (1) 支柱一：机器人控制 (Robot Control) (1) 支柱八：物理动画 (Physics-based Animation) (1)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (13 篇)

#题目一句话要点标签🔗⭐
1 Specialized Foundation Models Struggle to Beat Supervised Baselines 专业领域预训练大模型难胜监督学习基线模型 foundation model
2 Kolb-Based Experiential Learning for Generalist Agents with Human-Level Kaggle Data Science Performance Agent K：基于Kolb学习和Vygotsky ZPD的通用智能体，达到Kaggle数据科学人类水平 generalist agent
3 Long Context RAG Performance of Large Language Models 研究长上下文LLM在RAG中的性能，揭示其优势与局限性 large language model
4 Mobility-based Traffic Forecasting in a Multimodal Transport System 基于人口流动性的多模式交通系统流量预测研究 multimodal
5 Controlling for Unobserved Confounding with Large Language Model Classification of Patient Smoking Status 利用大型语言模型预测吸烟状态以控制未观察到的混杂因素 large language model
6 CE-CoLLM: Efficient and Adaptive Large Language Models Through Cloud-Edge Collaboration 提出CE-CoLLM云边协同框架，提升LLM在边缘环境的推理效率和适应性。 large language model ✅
7 Exploring Response Uncertainty in MLLMs: An Empirical Evaluation under Misleading Scenarios 揭示MLLM在误导信息下的响应不确定性，并提出MUB基准与微调策略 large language model multimodal ✅
8 GitChameleon: Unmasking the Version-Switching Capabilities of Code Generation Models 提出GitChameleon以解决代码生成模型版本适应性问题 large language model ✅
9 LASER: Attention with Exponential Transformation 提出LASER注意力机制，通过指数变换提升梯度信号，改善Transformer学习效率。 large language model
10 Climate AI for Corporate Decarbonization Metrics Extraction 提出CAI模型，利用LLM自动提取企业脱碳指标，提升数据收集效率和准确性。 large language model
11 DiffLM: Controllable Synthetic Data Generation via Diffusion Language Models DiffLM：通过扩散语言模型实现可控的合成数据生成 large language model ✅
12 Photon: Federated LLM Pre-Training Photon：首个端到端联邦LLM预训练系统，实现低带宽下的全局规模模型训练。 large language model
13 Stochastic Monkeys at Play: Random Augmentations Cheaply Break LLM Safety Alignment 随机增强可有效绕过大语言模型安全对齐，揭示其脆弱性 large language model ✅

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

#题目一句话要点标签🔗⭐
14 A Mamba Foundation Model for Time Series Forecasting 提出TSMamba，一种基于Mamba架构的时间序列预测线性复杂度基础模型。 Mamba foundation model
15 Layer-Adaptive State Pruning for Deep State Space Models 提出层自适应状态剪枝方法以优化深度状态空间模型 SSM state space model ✅
16 P-MOSS: Scheduling Main-Memory Indexes Over NUMA Servers Using Next Token Prediction P-MOSS：利用下一令牌预测在NUMA服务器上调度主存索引，提升查询吞吐量。 decision transformer large language model
17 On the Comparison between Multi-modal and Single-modal Contrastive Learning 通过信号噪声比分析，揭示多模态对比学习优于单模态对比学习的理论基础。 contrastive learning
18 A scalable generative model for dynamical system reconstruction from neuroimaging data 提出一种可扩展生成模型，用于从神经影像数据中重建动态系统。 SSM state space model
19 ADOPT: Modified Adam Can Converge with Any $β_2$ with the Optimal Rate 提出ADOPT以解决Adam优化算法收敛性问题 reinforcement learning deep reinforcement learning ✅

🔬 支柱五：交互与反应 (Interaction & Reaction) (1 篇)

#题目一句话要点标签🔗⭐
20 Privacy-Preserving Graph-Based Machine Learning with Fully Homomorphic Encryption for Collaborative Anti-Money Laundering 提出基于全同态加密的图机器学习方法，用于保护隐私的协同反洗钱。 OMOMO

🔬 支柱一：机器人控制 (Robot Control) (1 篇)

#题目一句话要点标签🔗⭐
21 Speech Separation with Pretrained Frontend to Minimize Domain Mismatch 提出自监督域不变预训练前端，缩小语音分离中真实数据与合成数据间的域差异 MPC

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

#题目一句话要点标签🔗⭐
22 Fast, robust approximate message passing 提出快速稳健的近似消息传递算法以解决优化问题 AMP

⬅️ 返回 cs.LG 首页 · 🏠 返回主页