cs.LG（2025-01-29）

📊 共 14 篇论文 | 🔗 2 篇有代码

🎯 兴趣领域导航

支柱九：具身大模型 (Embodied Foundation Models) (8) 支柱二：RL算法与架构 (RL & Architecture) (4 🔗1) 支柱一：机器人控制 (Robot Control) (1) 支柱四：生成式动作 (Generative Motion) (1 🔗1)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (8 篇)

#题目一句话要点标签🔗⭐
1 Current Pathology Foundation Models are unrobust to Medical Center Differences 揭示病理学Foundation Model对医学中心差异的非鲁棒性，提出鲁棒性指标。 foundation model
2 Topological Signatures of Adversaries in Multimodal Alignments 提出基于拓扑特征对比损失的多模态对抗攻击检测方法 multimodal
3 Fault Localization via Fine-tuning Large Language Models with Mutation Generated Stack Traces 通过微调大型语言模型与变异生成的堆栈跟踪来实现故障定位 large language model
4 Safeguarding Privacy in Edge Speech Understanding with Tiny Foundation Models 提出SpeechShield，利用微型语音模型在边缘设备上实现隐私保护的语音理解。 foundation model
5 AdditiveLLM: Large Language Models Predict Defects in Additive Manufacturing AdditiveLLM：利用大语言模型预测增材制造中的缺陷 large language model
6 DReSS: Data-driven Regularized Structured Streamlining for Large Language Models DReSS：数据驱动的正则化结构化精简方法，用于高效压缩大型语言模型 large language model
7 A Proximal Operator for Inducing 2:4-Sparsity 提出一种诱导2:4稀疏性的近端算子，提升大语言模型剪枝性能。 large language model
8 DFPE: A Diverse Fingerprint Ensemble for Enhancing LLM Performance 提出多样性指纹集成(DFPE)方法，提升LLM在复杂任务中的性能。 large language model

🔬 支柱二：RL算法与架构 (RL & Architecture) (4 篇)

#题目一句话要点标签🔗⭐
9 A Dual-Agent Adversarial Framework for Robust Generalization in Deep Reinforcement Learning 提出双智能体对抗框架，提升深度强化学习的泛化鲁棒性 reinforcement learning deep reinforcement learning policy learning
10 CAMP in the Odyssey: Provably Robust Reinforcement Learning with Certified Radius Maximization 提出CAMP：一种认证半径最大化的强化学习方法，提升对抗环境下的鲁棒性与回报。 reinforcement learning deep reinforcement learning DRL ✅
11 Temperature-Free Loss Function for Contrastive Learning 提出一种无温度超参数的对比学习损失函数，提升梯度特性并简化调参。 contrastive learning
12 ASAP: Learning Generalizable Online Bin Packing via Adaptive Selection After Proposal 提出ASAP框架，通过自适应选择提升在线装箱问题的泛化性和适应性。 reinforcement learning deep reinforcement learning DRL

🔬 支柱一：机器人控制 (Robot Control) (1 篇)

#题目一句话要点标签🔗⭐
13 From Sparse to Dense: Toddler-inspired Reward Transition in Goal-Oriented Reinforcement Learning 提出模仿幼儿学习的稀疏到稠密奖励过渡方法，提升强化学习效率 manipulation reinforcement learning egocentric

🔬 支柱四：生成式动作 (Generative Motion) (1 篇)

#题目一句话要点标签🔗⭐
14 Closing the Gap Between Synthetic and Ground Truth Time Series Distributions via Neural Mapping 提出NM-VQTSG以解决向量量化时间序列生成的保真度问题 VQ-VAE ✅

⬅️ 返回 cs.LG 首页 · 🏠 返回主页