cs.LG（2026-10-16）

📊 共 25 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) (13 🔗1) 支柱九：具身大模型 (Embodied Foundation Models) (9) 支柱三：空间感知与语义 (Perception & Semantics) (1) 支柱一：机器人控制 (Robot Control) (1) 支柱八：物理动画 (Physics-based Animation) (1)

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

#题目一句话要点标签🔗⭐
1 Tree-Aware Posterior Distillation for Long-Horizon Agentic Reinforcement Learning 提出TreeOPSD以解决长时间交互中的信用分配问题 reinforcement learning distillation large language model
2 Learning to Be Guided: Shaping Latent Geometry for Classifier-Free Guidance 提出原型正则化以优化无分类器引导生成性能 representation learning classifier-free guidance ✅
3 A Stronger Teacher Within: Strengthening MoE Models via On-Policy Bidirectional Distillation 提出MoEvolve框架以增强MoE模型的性能 reinforcement learning distillation privileged information
4 LED: Improving Masked Diffusion Language Models by Lookahead Editing 提出Lookahead Editing以优化掩码扩散语言模型 reinforcement learning MDM large language model
5 RECREG: Recovery-based Shielding for Regular Safety Properties in Reinforcement Learning 提出RecReg框架以解决强化学习中的安全性问题 reinforcement learning model-based RL
6 CocoRNA: Collective RNA Design with Cooperative Multi-agent Reinforcement Learning 提出CocoRNA以解决RNA序列设计的复杂性问题 reinforcement learning
7 JAXtari: High-Throughput and Easy-to-Modify Arcade Learning Environment 提出JAXtari以解决ALE环境修改与训练效率问题 reinforcement learning deep reinforcement learning PPO
8 Predictable Scaling: Efficient Model Exploration via Scaling-Law Prediction 提出可预测缩放框架以高效探索模型变更 MAE foundation model
9 Grounded Credit Assignment in Reinforcement Learning for Multi-Turn Search Agents 提出基于证据的信用分配方法以解决多轮搜索代理的奖励问题 reinforcement learning
10 Local proxies for memory-efficient training 提出局部代理网络以解决内存高效训练问题 distillation large language model
11 Path to Prior-Data Fitted Relational Models via Learned Latent Multi-table Representations 提出图基关系模型的合成预训练方法以提升泛化能力 curriculum learning foundation model
12 GR-OPD: Group-Relative Sequence Credit for On-Policy Distillation 提出GR-OPD以解决现有OPD方法的反馈信用分配问题 distillation
13 GRAFT-EEG: Gauge-Aware Physical Field Transport for Cross-Dataset Motor-Imagery EEG 提出GRAFT-EEG以解决跨数据集运动想象EEG解码问题 SSM state space model

🔬 支柱九：具身大模型 (Embodied Foundation Models) (9 篇)

#题目一句话要点标签🔗⭐
14 Domain-specific language alignment of multimodal embeddings towards explainable predictions 提出多模态嵌入的领域特定语言对齐框架以实现可解释预测 multimodal
15 Unit-level aggregation breaks coverage in causal foundation models 提出新方法以解决因果基础模型覆盖率下降问题 foundation model
16 Process Reward Modeling with Reliability- and Conservatism-Aware Bayesian Belief 提出BayesianPRM以解决过程奖励模型的不确定性问题 large language model multimodal
17 Universal Semantic Enhancement for Weakly Supervised Anomaly Detection 提出基于通用语义增强的弱监督异常检测方法以解决标签稀缺问题 foundation model
18 Expressivity of IQP Feature-Map Quantum Kernels via Reproducing Kernel Hilbert Spaces 提出IQP特征映射量子核以解决量子机器学习中的表达能力问题 multimodal
19 DeDiCa: A Scalable Deconfounding Discrete Causal Generative Model 提出DeDiCa以解决离散变量和隐藏混淆的因果推断问题 foundation model
20 Towards Hyperparameter Transfer for Differentially Private Optimization 提出DP-$μ$P以解决差分隐私优化中的超参数转移问题 zero-shot transfer
21 Large Diffusion Models Need More Than EMA 提出调度无关优化以提升大规模扩散模型训练效果 large language model
22 Benchmarking and Enhancing Agentic Causal Discovery 提出Heuresis基准以提升因果发现算法的性能 foundation model

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

#题目一句话要点标签🔗⭐
23 Deep LLM layers expose passage-level context in EEG during silent reading 提出深度LLM层以揭示静默阅读中的EEG段落级上下文 open-vocabulary open vocabulary large language model

🔬 支柱一：机器人控制 (Robot Control) (1 篇)

#题目一句话要点标签🔗⭐
24 Quality-of-Transmission Estimation in Quantum Networks: Compute, Simulate, or Learn? 提出计算、模拟或学习框架以优化量子网络的传输质量估计 sim-to-real

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

#题目一句话要点标签🔗⭐
25 HiPredict: Predicting the missing HiRISE RED4 spectral channel with machine learning 提出HiPredict以解决HiRISE RED4光谱通道缺失问题 spatiotemporal

⬅️ 返回 cs.LG 首页 · 🏠 返回主页