cs.LG（2026-03-31）

📊 共 20 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) (14 🔗2) 支柱二：RL算法与架构 (RL & Architecture) (4) 支柱一：机器人控制 (Robot Control) (1 🔗1) 支柱八：物理动画 (Physics-based Animation) (1)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (14 篇)

#	题目	一句话要点	标签	🔗	⭐
1	Aligned, Orthogonal or In-conflict: When can we safely optimize Chain-of-Thought?	提出CoT可监控性框架，预测训练如何影响LLM推理过程的可解释性	chain-of-thought
2	Quantifying Cross-Modal Interactions in Multimodal Glioma Survival Prediction via InterSHAP: Evidence for Additive Signal Integration	InterSHAP量化多模态胶质瘤生存预测中的交互作用，揭示加性信号整合机制。	multimodal
3	Real-Time Explanations for Tabular Foundation Models	提出ShapPFN以解决表格基础模型的可解释性问题	foundation model	✅
4	Multimodal Machine Learning for Early Prediction of Metastasis in a Swedish Multi-Cancer Cohort	提出一种多模态机器学习框架，用于提前一个月预测四种癌症的转移风险。	multimodal
5	Mind the Gap: A Framework for Assessing Pitfalls in Multimodal Active Learning	提出多模态主动学习评估框架，揭示现有方法在模态缺失和难度差异下的缺陷。	multimodal
6	Sampling at intermediate temperatures is optimal for training large language models in protein structure prediction	中间温度采样优化蛋白质结构预测中大型语言模型的训练	large language model
7	Survival In-Context: Prior-fitted In-context Learning Tabular Foundation Model for Survival Analysis	提出Survival In-Context，一种基于先验拟合的表格生存分析上下文学习基础模型	foundation model
8	A Comprehensive Information-Decomposition Analysis of Large Vision-Language Models	提出基于信息分解的LVLM分析框架，揭示多模态融合机制与模型策略。	multimodal	✅
9	Reward-Based Online LLM Routing via NeuralUCB	提出基于NeuralUCB的在线LLM路由方法，优化成本与奖励。	large language model
10	Think Anywhere in Code Generation	提出Think-Anywhere，解决代码生成中LLM推理时机不灵活问题	large language model
11	Task Scarcity and Label Leakage in Relational Transfer Learning	针对关系迁移学习中的任务稀缺和标签泄露问题，提出梯度投影方法抑制标签预测信息，提升模型泛化能力。	foundation model
12	Training-Free Dynamic Upcycling of Expert Language Models	提出DUME，无需训练即可动态整合专家语言模型，提升多领域性能。	large language model
13	One-for-All: A Lightweight Stabilized and Parameter-Efficient Pre-trained LLM for Time Series Forecasting	提出One-for-All框架，通过高斯秩稳定低秩适配器实现时间序列预测中预训练LLM的轻量化和参数高效微调。	large language model
14	Improving Ensemble Forecasts of Abnormally Deflecting Tropical Cyclones with Fused Atmosphere-Ocean-Terrain Data	提出AOT-TCs数据集和耦合模型，提升异常转向台风的集合预报精度	multimodal

🔬 支柱二：RL算法与架构 (RL & Architecture) (4 篇)

#	题目	一句话要点	标签	🔗	⭐
15	AP-DRL: A Synergistic Algorithm-Hardware Framework for Automatic Task Partitioning of Deep Reinforcement Learning on Versal ACAP	AP-DRL：用于Versal ACAP上深度强化学习自动任务划分的协同算法-硬件框架	reinforcement learning deep reinforcement learning DRL
16	Hybrid Quantum-Classical Spatiotemporal Forecasting for 3D Cloud Fields	提出QENO混合量子-经典时空预测框架，用于提升3D云场预测精度。	MAE spatiotemporal
17	Multi-AUV Cooperative Target Tracking Based on Supervised Diffusion-Aided Multi-Agent Reinforcement Learning	提出基于监督扩散辅助的多智能体强化学习算法，用于多AUV协同目标跟踪。	reinforcement learning policy learning
18	Target-Aligned Reinforcement Learning	提出目标对齐强化学习(TARL)，解决目标网络更新的稳定性-时效性权衡问题	reinforcement learning

🔬 支柱一：机器人控制 (Robot Control) (1 篇)

#	题目	一句话要点	标签	🔗	⭐
19	HCLSM: Hierarchical Causal Latent State Machines for Object-Centric World Modeling	HCLSM：用于对象中心世界建模的分层因果隐状态机	manipulation world model world models	✅

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

#	题目	一句话要点	标签	🔗	⭐
20	DiSGMM: A Method for Time-varying Microscopic Weight Completion on Road Networks	提出DiSGMM模型，用于补全路网中随时间变化的微观权重，提升交通态势感知。	spatiotemporal

⬅️ 返回 cs.LG 首页 · 🏠 返回主页