cs.LG（2026-09-15）

📊 共 20 篇论文 | 🔗 1 篇有代码

🎯 兴趣领域导航

支柱二：RL算法与架构 (RL & Architecture) (14 🔗1) 支柱九：具身大模型 (Embodied Foundation Models) (3) 支柱一：机器人控制 (Robot Control) (2) 支柱八：物理动画 (Physics-based Animation) (1)

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

#	题目	一句话要点	标签	🔗	⭐
1	OPD-Aha: From Linguistic Momentum to Visual Reflection in Multimodal On-Policy Distillation	提出OPD-Aha以解决多模态推理中的错误累积问题	teacher-student distillation multimodal	✅
2	SOTER: A Generative Time-Series Foundation Model for Wearable Human Physiological Signals	提出SOTER以解决可穿戴生理信号建模问题	MAE foundation model TAMP
3	Type-IV Code Clone Detection via Layer-Wise Non-Contrastive Representation Learning	提出LWVIC4Code以解决Type-IV代码克隆检测问题	representation learning contrastive learning large language model
4	Coupled Calibration and Learning: Mitigating Teacher Bias in LLM Distillation without Target-Domain Reward Feedback	提出Coupled Calibration and Learning以解决LLM蒸馏中的教师偏差问题	distillation large language model
5	Distributed JEPA: A Self-Supervised Framework for Energy Forecasting	提出分布式JEPA框架以解决能源预测问题	JEPA Joint-Embedding Predictive Architecture joint-embedding predictive architecture
6	Beyond Token-Local Imitation: Reward-Compatible Temporal Credit Assignment for On-Policy Distillation	提出γOPD以解决现有蒸馏方法的优化稳定性问题	distillation large language model
7	On the Importance of Gating: Memorization vs. In-Context Learning in State Space Models	探讨门控机制对状态空间模型的影响	SSM state space model
8	Conformal Policy Learning with Distribution-Free Safety Guarantees	提出共形政策学习以解决安全保障问题	policy learning
9	Same Flow, Different Paths: Variance Reduction in Flow Matching	提出路径选择优化以提高流匹配中的收敛速度	flow matching
10	Personalized Federated Learning through Global Knowledge Distillation and Local Head Adaptation	提出个性化联邦学习方法解决统计异质性问题	distillation
11	Repurposing Unified Topological Signatures for Graph Representation Learning	提出统一拓扑特征以提升图表示学习的能力	representation learning
12	HyCoSeq: Contextual Hyperbolic Representation Learning for Genomic Sequences	提出HyCoSeq框架以解决基因组序列表示学习问题	representation learning
13	GrowMTP: Can RL Grow Its Own Draft Head?	提出GrowMTP以解决RL训练中的草稿头生成问题	reinforcement learning large language model
14	What Does Layer-Importance Reveal About Transformers and State-Space Models?	通过层重要性分析揭示变换器与状态空间模型的差异	Mamba SSM

🔬 支柱九：具身大模型 (Embodied Foundation Models) (3 篇)

#	题目	一句话要点	标签	🔗	⭐
15	Large Language Models Develop Belief State Geometry In-Context	探讨大型语言模型的信念状态几何以提升上下文学习能力	large language model
16	ENCP: Episode-Normalized Conformal Prediction for Vision-and-Language Navigation	提出ENCP以解决视觉语言导航中的不确定性估计问题	vision-language-navigation VLN
17	Continuous-Time Machine Learning: A Unified Mathematical Perspective	提出统一数学视角以整合连续时间机器学习的主要分支	foundation model

🔬 支柱一：机器人控制 (Robot Control) (2 篇)

#	题目	一句话要点	标签	🔗	⭐
18	OptiPrime: Optimizing Private Inference through Protocol-Hardware Co-design	提出OptiPrime以优化私有推理中的协议与硬件协同设计	MPC OMOMO
19	LCAP: Population-Informed Latent Chip Adaptation from Few Output Probes for Photonic Neural Networks	提出LCAP框架以解决光子神经网络的硬件适应问题	sim-to-real

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

#	题目	一句话要点	标签	🔗	⭐
20	Recovering Physical Parameters from Fragmented Observations via Exact Distributed Spline Merging	通过精确分布样条合并恢复物理参数	spatiotemporal

⬅️ 返回 cs.LG 首页 · 🏠 返回主页