cs.LG（2025-06-06）

📊 共 35 篇论文 | 🔗 5 篇有代码

🎯 兴趣领域导航

支柱九：具身大模型 (Embodied Foundation Models) (17 🔗4) 支柱二：RL算法与架构 (RL & Architecture) (13) 支柱一：机器人控制 (Robot Control) (3 🔗1) 支柱八：物理动画 (Physics-based Animation) (1) 支柱三：空间感知与语义 (Perception & Semantics) (1)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (17 篇)

#题目一句话要点标签🔗⭐
1 MadaKV: Adaptive Modality-Perception KV Cache Eviction for Efficient Multimodal Long-Context Inference 提出MadaKV以解决多模态长上下文推理中的KV缓存效率问题 large language model multimodal
2 Multi-Modal Multi-Task Federated Foundation Models for Next-Generation Extended Reality Systems: Towards Privacy-Preserving Distributed Intelligence in AR/VR/MR 提出多模态多任务联邦基础模型以解决XR系统隐私问题 foundation model
3 BAQ: Efficient Bit Allocation Quantization for Large Language Models 提出BAQ以优化大语言模型的量化位分配问题 large language model ✅
4 Text-to-LoRA: Instant Transformer Adaption 提出Text-to-LoRA以解决大语言模型适应性问题 large language model foundation model ✅
5 Heartcare Suite: Multi-dimensional Understanding of ECG with Raw Multi-lead Signal Modeling 提出Heartcare Suite以解决心电图多维理解问题 large language model multimodal ✅
6 Dynamic Mixture of Progressive Parameter-Efficient Expert Library for Lifelong Robot Learning 提出动态混合渐进参数高效专家库以解决机器人终身学习问题 generalist agent
7 SPARQ: Synthetic Problem Generation for Reasoning via Quality-Diversity Algorithms 提出SPARQ以解决复杂数学问题生成的挑战 large language model
8 The Lock-in Hypothesis: Stagnation by Algorithm 提出锁定假说以解决算法引发的信念固化问题 large language model
9 Flexible Operator Fusion for Fast Sparse Transformer with Diverse Masking on GPU 提出STOF框架以优化稀疏Transformer的性能 large language model
10 LightGTS: A Lightweight General Time Series Forecasting Model 提出LightGTS以解决时间序列预测中的计算负担问题 foundation model
11 Mitigating Catastrophic Forgetting with Adaptive Transformer Block Expansion in Federated Fine-Tuning 提出FedBE以解决联邦微调中的灾难性遗忘问题 large language model
12 AQUATIC-Diff: Additive Quantization for Truly Tiny Compressed Diffusion Models 提出AQUATIC-Diff以解决扩散模型压缩问题 large language model
13 BestServe: Serving Strategies with Optimal Goodput in Collocation and Disaggregation Architectures 提出BestServe以优化大语言模型的服务策略 large language model
14 Training-Free Query Optimization via LLM-Based Plan Similarity 提出LLM-PM框架以实现无训练的查询优化 large language model
15 Come Together, But Not Right Now: A Progressive Strategy to Boost Low-Rank Adaptation 提出CoTo以解决低秩适应中的次优最小值问题 foundation model ✅
16 Contextually Guided Transformers via Low-Rank Adaptation 提出上下文引导变换器以解决提示依赖问题 large language model
17 Projectable Models: One-Shot Generation of Small Specialized Transformers from Large Ones 提出可投影模型以实现小型专用变换器的一次性生成 foundation model

🔬 支柱二：RL算法与架构 (RL & Architecture) (13 篇)

#题目一句话要点标签🔗⭐
18 BiTrajDiff: Bidirectional Trajectory Generation with Diffusion Models for Offline Reinforcement Learning 提出BiTrajDiff以解决离线强化学习中的数据分布偏差问题 reinforcement learning policy learning offline RL
19 How to craft a deep reinforcement learning policy for wind farm flow control 提出深度强化学习策略以优化风电场流动控制 reinforcement learning deep reinforcement learning
20 Debiasing Online Preference Learning via Preference Feature Preservation 提出偏好特征保留框架以解决在线偏好学习中的偏见问题 preference learning large language model
21 Delphos: A reinforcement learning framework for assisting discrete choice model specification 提出Delphos框架以优化离散选择模型的规范过程 reinforcement learning deep reinforcement learning
22 Reinforcement Learning Optimization for Large-Scale Learning: An Efficient and User-Friendly Scaling Library 提出ROLL库以解决大规模强化学习优化问题 reinforcement learning reward design
23 FlowOE: Imitation Learning with Flow Policy from Ensemble RL Experts for Optimal Execution under Heston Volatility and Concave Market Impacts 提出FlowOE以解决动态金融市场中的最优执行问题 imitation learning flow matching
24 Efficient Online RFT with Plug-and-Play LLM Judges: Unlocking State-of-the-Art Performance 提出高效在线RFT方法以解决RLHF中的奖励模型训练瓶颈 reinforcement learning PPO RLHF
25 Ensemble Elastic DQN: A novel multi-step ensemble approach to address overestimation in deep value-based reinforcement learning 提出EEDQN以解决深度强化学习中的过估计偏差问题 reinforcement learning deep reinforcement learning
26 Distillation Robustifies Unlearning 提出UNDO方法以增强大规模模型的去学习鲁棒性 distillation
27 Model-Driven Graph Contrastive Learning 提出MGCL以解决图对比学习中的数据增强问题 contrastive learning
28 Table-r1: Self-supervised and Reinforcement Learning for Program-based Table Reasoning in Small Language Models 提出Table-r1以解决小语言模型的表格推理问题 reinforcement learning
29 Exponential Family Variational Flow Matching for Tabular Data Generation 提出Exponential Family Variational Flow Matching以解决表格数据生成问题 flow matching
30 Learning Along the Arrow of Time: Hyperbolic Geometry for Backward-Compatible Representation Learning 提出超曲面几何方法以解决模型兼容性问题 representation learning

🔬 支柱一：机器人控制 (Robot Control) (3 篇)

#题目一句话要点标签🔗⭐
31 Gradual Transition from Bellman Optimality Operator to Bellman Operator in Online Reinforcement Learning 提出逐步过渡方法以提升在线强化学习的样本效率 locomotion manipulation reinforcement learning ✅
32 A Systematic Review of Poisoning Attacks Against Large Language Models 提出系统性框架以应对大型语言模型的中毒攻击问题 manipulation large language model
33 Physics-Informed Neural Networks for Control of Single-Phase Flow Systems Governed by Partial Differential Equations 提出物理信息神经网络以控制单相流动系统 MPC model predictive control

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

#题目一句话要点标签🔗⭐
34 Integrating Spatiotemporal Features in LSTM for Spatially Informed COVID-19 Hospitalization Forecasting 提出并行流LSTM框架以提升COVID-19住院预测准确性 spatiotemporal

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

#题目一句话要点标签🔗⭐
35 NeurNCD: Novel Class Discovery via Implicit Neural Representation 提出NeurNCD以解决开放世界中新类发现问题 NeRF

⬅️ 返回 cs.LG 首页 · 🏠 返回主页