| 1 |
Empowering Source-Free Domain Adaptation via MLLM-Guided Reliability-Based Curriculum Learning |
提出基于MLLM指导的可靠性课程学习，解决无源域自适应问题 |
curriculum learning large language model foundation model |
✅ |
|
| 2 |
HarmoDT: Harmony Multi-Task Decision Transformer for Offline Reinforcement Learning |
HarmoDT：通过和谐参数子空间学习，解决离线多任务强化学习中的策略优化问题 |
reinforcement learning offline reinforcement learning decision transformer |
|
|
| 3 |
Large Language Model-Driven Curriculum Design for Mobile Networks |
提出基于大语言模型的移动网络课程设计框架，提升强化学习性能 |
reinforcement learning curriculum learning large language model |
|
|
| 4 |
Reinforcement Learning in Dynamic Treatment Regimes Needs Critical Reexamination |
重新评估动态治疗方案中离线强化学习的应用有效性 |
reinforcement learning offline RL offline reinforcement learning |
✅ |
|
| 5 |
SleepFM: Multi-modal Representation Learning for Sleep Across Brain Activity, ECG and Respiratory Signals |
SleepFM：通过脑电、心电和呼吸信号的多模态表征学习用于睡眠分析 |
representation learning contrastive learning foundation model |
✅ |
|
| 6 |
Atlas3D: Physically Constrained Self-Supporting Text-to-3D for Simulation and Fabrication |
Atlas3D：物理约束的自支撑文本到3D生成，用于仿真和制造 |
distillation differentiable simulation embodied AI |
|
|
| 7 |
Bridging Mini-Batch and Asymptotic Analysis in Contrastive Learning: From InfoNCE to Kernel-Based Losses |
对比学习中，从InfoNCE到核方法的损失函数统一性分析与新损失函数DHEL的提出 |
representation learning contrastive learning |
|
|
| 8 |
Offline-Boosted Actor-Critic: Adaptively Blending Optimal Historical Behaviors in Deep Off-Policy RL |
提出离线增强的Actor-Critic算法，自适应融合历史最优行为以提升深度离线策略强化学习性能。 |
reinforcement learning policy learning offline RL |
|
|
| 9 |
In-Context Symmetries: Self-Supervised Learning through Contextual World Models |
提出ContextSSL，通过上下文世界模型自监督学习任务自适应的对称性表示。 |
world model |
|
|
| 10 |
No $D_{\text{train}}$: Model-Agnostic Counterfactual Explanations Using Reinforcement Learning |
提出NTD-CFE，一种无需训练数据的模型无关强化学习反事实解释方法，适用于静态和时序数据。 |
reinforcement learning |
|
|
| 11 |
Highway Reinforcement Learning |
提出高架门以解决多步离策略强化学习中的低估问题 |
reinforcement learning |
|
|
| 12 |
Back to the Drawing Board for Fair Representation Learning |
重新审视公平表征学习：关注迁移任务以避免过拟合代理任务 |
representation learning |
|
|
| 13 |
Individual Contributions as Intrinsic Exploration Scaffolds for Multi-agent Reinforcement Learning |
提出ICES，利用个体贡献作为内在探索支架，解决MARL中的稀疏奖励探索问题。 |
reinforcement learning |
✅ |
|
| 14 |
A Pontryagin Perspective on Reinforcement Learning |
提出基于庞特里亚金原理的开环强化学习算法，提升高维控制任务性能 |
reinforcement learning |
|
|
| 15 |
Mutation-Bias Learning in Games |
提出基于演化博弈论的突变偏差多智能体强化学习算法，提升复杂环境收敛性。 |
reinforcement learning PHC |
|
|
| 16 |
Mollification Effects of Policy Gradient Methods |
揭示策略梯度方法对非光滑优化问题的平滑效应及其局限性 |
reinforcement learning deep reinforcement learning |
|
|
| 17 |
AlignIQL: Policy Alignment in Implicit Q-Learning through Constrained Optimization |
AlignIQL：通过约束优化在隐式Q学习中实现策略对齐 |
offline RL IQL |
|
|
| 18 |
Imitating from auxiliary imperfect demonstrations via Adversarial Density Weighted Regression |
提出对抗密度加权回归(ADR)模仿学习框架，利用辅助数据提升策略性能。 |
IQL imitation learning |
✅ |
|