cs.LG（2026-08-25）

📊 共 21 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) (12 🔗1) 支柱九：具身大模型 (Embodied Foundation Models) (7) 支柱三：空间感知与语义 (Perception & Semantics) (1) 支柱一：机器人控制 (Robot Control) (1)

🔬 支柱二：RL算法与架构 (RL & Architecture) (12 篇)

#题目一句话要点标签🔗⭐
1 XP-JEPA: Cross-Predictive Physics Grounding for Forecastable Latent Dynamics 提出XP-JEPA以解决自预测模型的物理约束不足问题 world model world models JEPA
2 CoDrift: Compositional Drifting for Offline Reinforcement Learning 提出CoDrift以解决离线强化学习中的多目标问题 reinforcement learning policy learning offline RL
3 Bellman Calibration for Marginalized Importance Weighting in Offline Reinforcement Learning 提出同调贝尔曼校准以解决离线强化学习中的重要性加权问题 reinforcement learning offline reinforcement learning
4 On-policy Distillation with Verifiable Reward 提出OPDVR以解决稀疏反馈与轨迹正确性问题 reinforcement learning distillation large language model ✅
5 Robust Data-Collection Policy Learning for Low-Variance Online Policy Evaluation 提出双循环梯度算法以解决强化学习中的高方差评估问题 reinforcement learning policy learning
6 IAPO: Influence-Aware Policy Optimization for Credit Assignment in Multi-Turn Service Agents 提出影响感知策略优化以解决多轮服务代理中的信用分配问题 reinforcement learning large language model
7 SeisMamba: Low-Latency Single-Station Seismic Magnitude Estimation for Spatially Distributed Earthquake Early Warning 提出SeisMamba以解决单站地震震级快速估计问题 Mamba
8 Where Entropy Is Measured Matters: Policy Geometry in Bounded Continuous-Control PPO 提出改进的PPO方法以优化连续控制策略的熵测量 PPO
9 Evaluating Deep Multivariate Imputation Models on Wearable Device Data 提出深度多变量插补模型评估方法以解决可穿戴设备数据缺失问题 MAE multimodal
10 PhysicsBench: A Unified Leaderboard for Generative and Predictive Models in Engineering Design and Simulation 提出PhysicsBench以统一评估工程设计中的生成与预测模型 predictive model
11 ChorusTIC: Training-Free Multivariate Time Series Classification via Chorus In-Context Learning 提出ChorusTIC以解决多变量时间序列分类问题 representation learning foundation model
12 Low-Latency Activation-Regularized Sparse Neural Operators with Distillation Assistance Towards Real-Time Edge-Deployable Virtual Sensing 提出稀疏激活正则化神经算子以解决边缘部署的实时虚拟传感问题 distillation

🔬 支柱九：具身大模型 (Embodied Foundation Models) (7 篇)

#题目一句话要点标签🔗⭐
13 A Multimodal Foundation Model for Longitudinal Patient Representation and Scalable Insight Generation in Oncology 提出多模态基础模型以解决肿瘤学中的患者长期状态表示问题 foundation model multimodal
14 LION: A Clifford Neural Paradigm for Multimodal-Attributed Graph Learning 提出LION以解决多模态图学习中的对齐与融合问题 multimodal
15 Taming foundation model with invariance-oriented pre-training for broad-spectrum EEG analysis across signal-level, brain-state, and brain-health tasks 提出INCEPT模型以解决EEG分析中的可重用性问题 foundation model
16 Causal Analysis for Time Series Foundation Models 提出因果分析框架以识别时间序列基础模型的偏差与失败模式 foundation model
17 Data Leakage Inflates Generalizability of Power Outage Prediction Models 揭示数据泄露对电力中断预测模型泛化能力的影响 foundation model
18 When Do Supervised UQ Ensembles Improve LLM Hallucination Detection? A Robustness Study 提出监督式集成方法以提升LLM幻觉检测的鲁棒性 large language model
19 From Gradient-Boosted Trees to Deep Recommenders: Practical Lessons from Migrating a Production Customer Support Recommender 提出深度推荐系统以解决传统树模型在动态推荐中的局限性 multimodal

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

#题目一句话要点标签🔗⭐
20 Physics-Integrated Operator Learning via Gaussian Splatting Representations 提出基于高斯点云表示的物理集成算子学习方法 gaussian splatting splatting spatiotemporal

🔬 支柱一：机器人控制 (Robot Control) (1 篇)

#题目一句话要点标签🔗⭐
21 WarpSAC: Towards the Pinnacle of Scalable Off-policy RL by Rethinking Exploration and Exploitation 提出WarpSAC以解决可扩展离线强化学习中的探索与利用问题 manipulation sim-to-real Unitree

⬅️ 返回 cs.LG 首页 · 🏠 返回主页