cs.LG（2026-08-20）

📊 共 18 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) (6) 支柱九：具身大模型 (Embodied Foundation Models) (6 🔗1) 支柱四：生成式动作 (Generative Motion) (2) 支柱七：动作重定向 (Motion Retargeting) (1 🔗1) 支柱三：空间感知与语义 (Perception & Semantics) (1) 支柱八：物理动画 (Physics-based Animation) (1) 支柱一：机器人控制 (Robot Control) (1)

🔬 支柱二：RL算法与架构 (RL & Architecture) (6 篇)

#	题目	一句话要点	标签	🔗	⭐
1	Orthogonal JEPA: Factorized Predictive States for Latent World Models	提出正交JEPA以解决复杂系统中的预测状态问题	world model world models JEPA
2	FAR-DPO: Feasibility-Aware and Robust Direct Preference Optimization for Cyclic Peptide Design	提出FAR-DPO以解决循环肽设计中的可行性与鲁棒性问题	DPO direct preference optimization
3	Adaptive Probabilistic Shielding by Learning MDPs for Safe Reinforcement Learning	提出自适应概率屏蔽以解决安全强化学习问题	reinforcement learning
4	Truncate Bad, Upweight Good: BoN-Style Distillation via Rank-Based Classification	提出TUP策略以解决低排名样本影响的问题	distillation
5	When Graph-JEPA Learns the Wrong Thing: Diagnosing and Repairing Category-Conditional Collapse	提出Graph-JEPA修复类别条件崩溃问题的诊断方法	JEPA Joint-Embedding Predictive Architecture joint-embedding predictive architecture
6	aiXamine: Unified Black-Box Evaluation of Cross-Dimensional Trade-offs in LLM Safety, Security, and Privacy	提出aiXamine以解决LLM安全性、隐私和安全性评估的多维问题	distillation large language model

🔬 支柱九：具身大模型 (Embodied Foundation Models) (6 篇)

#	题目	一句话要点	标签	🔗	⭐
7	Scale-Aware Pretraining of Time Series Foundation Models via Multi-Patch Token Alignment and Hybrid Masking	提出SATS以解决时间序列预训练中的尺度不一致问题	foundation model
8	DraftFM: A FoundationModel for Day-Zero Drafting in Magic: The Gathering	提出DraftFM以解决Magic: The Gathering的日零草拟问题	foundation model
9	Let's Scale Step by Step: Compute-Efficient Hyperparameter Transfer for Large-Scale Mixture-of-Experts	提出计算高效的超参数转移框架以优化大规模MoE模型	foundation model
10	Systematic Evaluation of TabPFN-TS for Zero-Shot Probabilistic Heat Load Forecasting in District Heating Networks	提出TabPFN-TS以解决区域供热网络的零-shot热负荷预测问题	foundation model
11	RecPFN: Prior-Fitted Networks for In-Context-Based Recommendations	提出RecPFN以解决序列推荐中的数据效率问题	multimodal	✅
12	AgentDecarbonizer: Carbon-Aware Execution for AI Agents	提出AgentDecarbonizer以解决AI代理的碳排放问题	large language model

🔬 支柱四：生成式动作 (Generative Motion) (2 篇)

#	题目	一句话要点	标签	🔗	⭐
13	A Locally Tokenized Generative Model for Robust Time-Series Watermarking	提出L-VQVAE与LVQMark以解决多变量时间序列水印的可靠性问题	VQ-VAE
14	Wrong-Physics Backdoors in Neural PDE Operators	提出错误物理后门以解决神经偏微分方程算子的验证问题	physically plausible

🔬 支柱七：动作重定向 (Motion Retargeting) (1 篇)

#	题目	一句话要点	标签	🔗	⭐
15	G-MARK: Grounded Multi-Agent Reasoning for Cooperative Driving via Knowledge Graphs	提出G-MARK以解决自主驾驶中的多代理推理问题	motion prediction multimodal	✅

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

#	题目	一句话要点	标签	🔗	⭐
16	Decoding silent reading from non-invasive EEG	通过无创EEG解码静默阅读中的词汇信息	open-vocabulary open vocabulary large language model

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

#	题目	一句话要点	标签	🔗	⭐
17	A comparison between ceiling-mounted FMCW, IR-UWB and Wi-Fi radar for in-bedroom human activity monitoring and sleep interruption detection	比较天花板安装的FMCW、IR-UWB和Wi-Fi雷达以监测卧室人类活动和睡眠干扰	PULSE

🔬 支柱一：机器人控制 (Robot Control) (1 篇)

#	题目	一句话要点	标签	🔗	⭐
18	Keyed Provenance Watermarking with Complementary Lattice-Based Secure Aggregation for Federated Learning	提出基于水印和格基安全聚合的联邦学习框架以应对多重攻击	manipulation Mamba linear attention

⬅️ 返回 cs.LG 首页 · 🏠 返回主页