cs.LG（2025-06-05）

📊 共 38 篇论文 | 🔗 8 篇有代码

🎯 兴趣领域导航

支柱九：具身大模型 (Embodied Foundation Models) (20 🔗2) 支柱二：RL算法与架构 (RL & Architecture) (14 🔗5) 支柱一：机器人控制 (Robot Control) (1 🔗1) 支柱七：动作重定向 (Motion Retargeting) (1) 支柱五：交互与反应 (Interaction & Reaction) (1) 支柱八：物理动画 (Physics-based Animation) (1)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (20 篇)

#题目一句话要点标签🔗⭐
1 Tuning the Right Foundation Models is What you Need for Partial Label Learning 提出PartialCLIP以解决部分标签学习中的模型选择问题 foundation model ✅
2 PCDVQ: Enhancing Vector Quantization for Large Language Models via Polar Coordinate Decoupling 提出PCDVQ以解决大语言模型量化精度不足问题 large language model
3 LSM-2: Learning from Incomplete Wearable Sensor Data 提出LSM-2以解决可穿戴传感器数据不完整问题 foundation model multimodal
4 Scaling Laws for Robust Comparison of Open Foundation Language-Vision Models and Datasets 提出缩放法则以比较语言-视觉模型与数据集 foundation model ✅
5 Conformal Prediction Adaptive to Unknown Subpopulation Shifts 提出适应未知子群体转变的保形预测方法 large language model
6 Ravan: Multi-Head Low-Rank Adaptation for Federated Fine-Tuning 提出Ravan以解决联邦微调中的低秩适应问题 large language model
7 Conformal Prediction Beyond the Seen: A Missing Mass Perspective for Uncertainty Quantification in Generative Models 提出CPQ框架以解决生成模型的不确定性量化问题 large language model
8 Power Law Guided Dynamic Sifting for Efficient Attention 提出SiftAttention以解决GPU上大语言模型的内存带宽限制问题 large language model
9 Sample Complexity and Representation Ability of Test-time Scaling Paradigms 提出测试时缩放范式以提升大语言模型的样本效率 large language model
10 Transformers Meet In-Context Learning: A Universal Approximation Theory 提出通用逼近理论以解释变换器的上下文学习能力 large language model
11 Membership Inference Attacks on Sequence Models 提出基于序列模型的成员推断攻击以提高隐私审计效果 large language model
12 QiMeng: Fully Automated Hardware and Software Design for Processor Chip 提出QiMeng以实现处理器芯片的全自动硬件和软件设计 large language model
13 BacPrep: An Experimental Platform for Evaluating LLM-Based Bacalaureat Assessment 提出BacPrep平台以解决罗马尼亚高考备考反馈不足问题 large language model
14 FPTQuant: Function-Preserving Transforms for LLM Quantization 提出FPTQuant以解决大语言模型量化效率问题 large language model
15 Agentic AI for Intent-Based Industrial Automation 提出意图驱动的Agentic AI框架以简化工业自动化 large language model
16 Sparse Autoencoders, Again? 提出混合模型以解决稀疏自编码器的局限性 large language model
17 Enhancing Delta Compression in LLMs via SVD-based Quantization Error Minimization 提出DeltaMix框架以解决LLMs的量化误差问题 large language model
18 Towards Better Generalization via Distributional Input Projection Network 提出分布式输入投影网络以提升模型泛化能力 large language model
19 MobiEdit: Resource-efficient Knowledge Editing for Personalized On-device LLMs 提出MobiEdit以解决移动设备上个性化LLM知识编辑问题 large language model
20 Clustering and Median Aggregation Improve Differentially Private Inference 通过聚类与中位数聚合提升差分隐私推断质量 large language model

🔬 支柱二：RL算法与架构 (RL & Architecture) (14 篇)

#题目一句话要点标签🔗⭐
21 Causal Policy Learning in Reinforcement Learning: Backdoor-Adjusted Soft Actor-Critic 提出DoSAC以解决强化学习中的隐性混淆问题 reinforcement learning policy learning SAC
22 Dissecting Long-Chain-of-Thought Reasoning Models: An Empirical Study 系统分析长链推理模型以提升推理能力与效率 reinforcement learning chain-of-thought ✅
23 Aligning Multimodal Representations through an Information Bottleneck 通过信息瓶颈原理提出新方法以解决多模态表示对齐问题 representation learning multimodal
24 TabFlex: Scaling Tabular Learning to Millions with Linear Attention 提出TabFlex以解决大规模表格学习效率问题 linear attention large language model
25 Agentomics-ML: Autonomous Machine Learning Experimentation Agent for Genomic and Transcriptomic Data 提出Agentomics-ML以解决生物数据自动化建模问题 predictive model large language model multimodal ✅
26 Mixture-of-Experts Meets In-Context Reinforcement Learning 提出T2MIR框架以解决ICRL中的多模态与任务异质性问题 reinforcement learning contrastive learning ✅
27 StatsMerging: Statistics-Guided Model Merging via Task-Specific Teacher Distillation 提出StatsMerging以解决模型合并中的标签依赖问题 distillation
28 Two-dimensional Taxonomy for N-ary Knowledge Representation Learning Methods 提出二维分类法以解决n元知识表示学习的复杂性问题 representation learning
29 Improving Data Efficiency for LLM Reinforcement Fine-tuning Through Difficulty-targeted Online Data Selection and Rollout Replay 提出难度针对的在线数据选择与回放重放以提高LLM强化微调的数据效率 reinforcement learning large language model ✅
30 Mitigating Degree Bias Adaptively with Hard-to-Learn Nodes in Graph Contrastive Learning 提出HAR损失以适应性缓解图对比学习中的度偏差问题 contrastive learning
31 TreeRPO: Tree Relative Policy Optimization 提出TreeRPO以优化推理过程中的奖励信号 reinforcement learning large language model ✅
32 UnHiPPO: Uncertainty-aware Initialization for State Space Models 提出UnHiPPO以解决状态空间模型中的噪声问题 state space model
33 When Maximum Entropy Misleads Policy Optimization 分析最大熵强化学习在控制任务中的误导性 reinforcement learning reward design
34 Learning long range dependencies through time reversal symmetry breaking 提出RHEL算法以解决长程依赖学习问题 SSM state space model

🔬 支柱一：机器人控制 (Robot Control) (1 篇)

#题目一句话要点标签🔗⭐
35 A Smooth Sea Never Made a Skilled SAILOR: Robust Imitation via Learning to Search 提出SAILOR以解决行为克隆方法的局限性 manipulation imitation learning diffusion policy ✅

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

#题目一句话要点标签🔗⭐
36 Multi-Point Proximity Encoding For Vector-Mode Geospatial Machine Learning 提出多点接近编码以解决向量模式地理空间机器学习问题 spatial relationship

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

#题目一句话要点标签🔗⭐
37 Spectral Graph Neural Networks are Incomplete on Graphs with a Simple Spectrum 提出旋转等变神经网络以提升谱图神经网络的表达能力 OMOMO

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

#题目一句话要点标签🔗⭐
38 FaCTR: Factorized Channel-Temporal Representation Transformers for Efficient Time Series Forecasting 提出FaCTR以解决时间序列预测中的过度参数化问题 spatiotemporal

⬅️ 返回 cs.LG 首页 · 🏠 返回主页