cs.LG（2026-06-17）

📊 共 28 篇论文 | 🔗 4 篇有代码

🎯 兴趣领域导航

支柱二：RL算法与架构 (RL & Architecture) (11 🔗1) 支柱九：具身大模型 (Embodied Foundation Models) (10 🔗3) 支柱一：机器人控制 (Robot Control) (3) 支柱三：空间感知与语义 (Perception & Semantics) (2) 支柱八：物理动画 (Physics-based Animation) (1) 支柱五：交互与反应 (Interaction & Reaction) (1)

🔬 支柱二：RL算法与架构 (RL & Architecture) (11 篇)

#题目一句话要点标签🔗⭐
1 Seeing Before Reasoning: Decoupling Perception and Reasoning for Shortcut-Resilient Multimodal On-Policy Self-Distillation 提出ViGOS框架以解决多模态自蒸馏中的快捷问题 distillation large language model multimodal
2 Dual-Channel Grounded World Modeling (DCGWM): Structural Prevention of Objective Interference Collapse via Heterogeneous External Grounding with Inward-Only Gradient Flow 提出双通道基础世界建模以解决目标干扰崩溃问题 world model world models JEPA
3 Reinforcement Learning Foundation Models Should Already Be A Thing 提出强化学习基础模型以填补现有方法空白 reinforcement learning foundation model
4 The Reward Was in Your Data All Along: Correcting Flow Matching with Discriminator-Guided RL 提出判别器引导的强化学习以解决流匹配问题 reinforcement learning DRL flow matching
5 Learning from Your Own Mistakes: Constructing Learnable Micro-Reflective Trajectories for Self-Distillation 提出TAPO以解决自蒸馏中的错误诊断问题 distillation large language model
6 Learning from Own Solutions: Self-Conditioned Credit Assignment for Reinforcement Learning with Verifiable Rewards 提出SC-GRPO以解决强化学习中的信用分配问题 reinforcement learning distillation privileged information
7 UBP2: Uncertainty-Balanced Preference Planning for Efficient Preference-based Reinforcement Learning 提出UBP2以解决偏好强化学习中的样本效率问题 reinforcement learning reward design
8 Wasserstein Policy Learning for Distributional Outcomes 提出Wasserstein策略学习以解决离线政策学习中的分布值结果问题 policy learning
9 REVES: REvision and VErification--Augmented Training for Test-Time Scaling 提出REVES以解决大语言模型推理中的多步骤优化问题 reinforcement learning large language model ✅
10 Lifecycle-Aware Dynamic Analysis for Secure ML Model Execution 提出生命周期感知动态分析以保障机器学习模型执行安全 world model world models
11 Maturing Markov Decision Processes: Decision Making under Increasing Information and Shrinking Action Sets 提出成熟马尔可夫决策过程以解决信息不对称问题 reinforcement learning distillation

🔬 支柱九：具身大模型 (Embodied Foundation Models) (10 篇)

#题目一句话要点标签🔗⭐
12 Does VLA Even Know the Basics? Measuring Commonsense and World Knowledge Retention in Vision-Language-Action Models 提出Act2Answer以评估VLA模型的常识与知识保留能力 vision-language-action VLA ✅
13 ChronoSurv: A Clinical Pathway-Guided Graph Framework for Multimodal Survival Analysis 提出ChronoSurv以解决多模态生存分析中的临床路径问题 multimodal
14 Robust and Interpretable Adaptation of Equivariant Materials Foundation Models via Sparsity-promoting Fine-tuning 提出稀疏促进微调方法以解决材料基础模型适应性问题 foundation model
15 Bounded Context Management for Tabular Foundation Models on Stream Learning 提出CURE以解决表格流学习中的上下文管理问题 foundation model ✅
16 Structured Inference with Large Language Gibbs 提出Large Language Gibbs以解决结构化推理问题 large language model
17 OpenAnt: LLM-Powered Vulnerability Discovery Through Code Decomposition, Adversarial Verification, and Dynamic Testing 提出OpenAnt以解决大规模代码库中的自动化漏洞发现问题 large language model ✅
18 Quantifying and Auditing LLM Evaluation via Positive--Unlabeled Learning 提出正负样本学习框架以审计LLM评估偏差 large language model
19 FoMoE: Breaking the Full-Replica Barrier with a Federation of MoEs 提出FoMoE以打破全复制限制，实现专家层分布式训练 large language model
20 Graph Grounded Cross Attention Transformer Neural Network for Structurally Constrained Full Event Sequence Generation in Predictive Process Monitoring 提出图基础交叉注意力变换器网络以解决结构约束事件序列生成问题 TAMP
21 BLADE: Scalable Bi-level Adaptive Data Selection for LLM Training 提出BLADE以解决大规模LLM训练中的数据选择问题 large language model

🔬 支柱一：机器人控制 (Robot Control) (3 篇)

#题目一句话要点标签🔗⭐
22 Stealthy World Model Manipulation via Data Poisoning 提出SWAAP框架以解决世界模型数据中毒问题 manipulation world model world models
23 Giskard : Byzantine Robust and Confidential Aggregation for Large-Scale Decentralized Learning 提出Giskard以解决去中心化学习中的保密性与拜占庭问题 MPC
24 Strategic Feature Selection 提出战略特征选择方法以应对算法决策中的操控问题 manipulation

🔬 支柱三：空间感知与语义 (Perception & Semantics) (2 篇)

#题目一句话要点标签🔗⭐
25 Point-Cloud-Assistant Localized Statistical Channel Prediction by Tangent Gaussian Splatting 提出点云辅助的切线高斯喷溅以解决无线网络信道预测问题 gaussian splatting splatting
26 Semantic Robustness Certification for Vision-Language Models 提出语义鲁棒性认证框架以解决视觉语言模型的鲁棒性问题 open-vocabulary open vocabulary

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

#题目一句话要点标签🔗⭐
27 P-K-GCN: Physics-augmented Koopman-enhanced Graph Convolutional Network for Deep Spatiotemporal Super-resolution 提出P-K-GCN以解决不规则几何体的时空超分辨率问题 spatiotemporal

🔬 支柱五：交互与反应 (Interaction & Reaction) (1 篇)

#题目一句话要点标签🔗⭐
28 Diffusion-Proof: Recipe for Formal Theorem Proving Beyond Auto-Regressive Generation 提出Diffusion-Proof以解决形式定理证明中的长程一致性问题 IMoS large language model

⬅️ 返回 cs.LG 首页 · 🏠 返回主页