cs.LG（2025-11-07）

📊 共 20 篇论文 | 🔗 2 篇有代码

🎯 兴趣领域导航

支柱二：RL算法与架构 (RL & Architecture) (9 🔗1) 支柱九：具身大模型 (Embodied Foundation Models) (8) 支柱八：物理动画 (Physics-based Animation) (2 🔗1) 支柱三：空间感知与语义 (Perception & Semantics) (1)

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

#题目一句话要点标签🔗⭐
1 An End-to-End Deep Reinforcement Learning Approach for Solving the Traveling Salesman Problem with Drones 提出基于Transformer和Minimal Gated Unit的深度强化学习框架，解决带无人机的旅行商问题。 reinforcement learning deep reinforcement learning policy learning
2 You Need Reasoning to Learn Reasoning: The Limitations of Label-Free RL in Weak Base Models 研究表明，无监督强化学习提升小模型推理能力受限，并提出课程学习与数据筛选方法。 reinforcement learning curriculum learning large language model ✅
3 A Metamorphic Testing Perspective on Knowledge Distillation for Language Models of Code: Does the Student Deeply Mimic the Teacher? 提出MetaCompress，用于评估代码语言模型知识蒸馏的保真度。 distillation large language model
4 Multi-agent Coordination via Flow Matching 提出MAC-Flow，通过流匹配实现高效多智能体协同 reinforcement learning flow matching
5 MedFedPure: A Medical Federated Framework with MAE-based Detection and Diffusion Purification for Inference-Time Attacks MedFedPure：一种基于MAE检测和扩散净化的医学联邦学习框架，用于防御推理时攻击。 masked autoencoder MAE
6 Sample Complexity of Distributionally Robust Off-Dynamics Reinforcement Learning with Online Interaction 提出基于在线交互的分布鲁棒离策略强化学习算法，解决探索难题。 reinforcement learning
7 SAD-Flower: Flow Matching for Safe, Admissible, and Dynamically Consistent Planning SAD-Flower：通过流匹配实现安全、容许和动态一致的规划 flow matching
8 On Flow Matching KL Divergence 推导Flow Matching KL散度的确定性上界，提升生成模型统计效率 flow matching
9 Distributionally Robust Self Paced Curriculum Reinforcement Learning 提出DR-SPCRL，通过自步课程学习调整鲁棒性预算，提升强化学习在分布偏移下的性能。 reinforcement learning

🔬 支柱九：具身大模型 (Embodied Foundation Models) (8 篇)

#题目一句话要点标签🔗⭐
10 Turning Adversaries into Allies: Reversing Typographic Attacks for Multimodal E-Commerce Product Retrieval 提出一种反转印刷攻击的视觉-文本压缩方法，提升电商多模态商品检索性能。 foundation model multimodal
11 Model Merging Improves Zero-Shot Generalization in Bioacoustic Foundation Models 模型融合提升生物声学基础模型零样本泛化能力 foundation model instruction following
12 Distributionally Robust Multimodal Machine Learning 提出分布鲁棒多模态学习框架，提升不确定性下的多模态融合性能。 multimodal
13 Multimodal Deep Learning for Prediction of Progression-Free Survival in Patients with Neuroendocrine Tumors Undergoing 177Lu-based Peptide Receptor Radionuclide Therapy 多模态深度学习预测177Lu-PRRT治疗神经内分泌肿瘤患者的无进展生存期 multimodal
14 Beyond Redundancy: Diverse and Specialized Multi-Expert Sparse Autoencoder 提出多样化专家混合稀疏自编码器，提升大语言模型的可解释性与效率。 large language model
15 Usando LLMs para Programar Jogos de Tabuleiro e Variações 利用大型语言模型辅助棋盘游戏及其变种的程序开发 large language model
16 OvA-LP: A Simple and Efficient Framework for Federated Learning on Non-IID Data OvA-LP：一种简单高效的联邦学习框架，用于解决非独立同分布数据下的模型漂移问题 foundation model
17 Leak@$k$: Unlearning Does Not Make LLMs Forget Under Probabilistic Decoding 揭示LLM在概率解码下遗忘失效：提出Leak@$k$评估知识泄漏 large language model

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

#题目一句话要点标签🔗⭐
18 Evaluating Spatio-Temporal Forecasting Trade-offs Between Graph Neural Networks and Foundation Models 评估图神经网络与时间序列基础模型在时空预测中的权衡，优化传感器部署。 spatiotemporal foundation model ✅
19 SSTODE: Ocean-Atmosphere Physics-Informed Neural ODEs for Sea Surface Temperature Prediction SSTODE：基于海洋-大气物理信息的神经ODE，用于海面温度预测 spatiotemporal

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

#题目一句话要点标签🔗⭐
20 Precipitation nowcasting of satellite data using physically-aligned neural networks 提出TUPANN，一种基于物理对齐神经网络的卫星降水临近预报模型 optical flow

⬅️ 返回 cs.LG 首页 · 🏠 返回主页