cs.LG（2023-10-05）

📊 共 22 篇论文 | 🔗 5 篇有代码

🎯 兴趣领域导航

支柱九：具身大模型 (Embodied Foundation Models) (10 🔗3) 支柱二：RL算法与架构 (RL & Architecture) (8 🔗2) 支柱一：机器人控制 (Robot Control) (2) 支柱三：空间感知与语义 (Perception & Semantics) (1) 支柱八：物理动画 (Physics-based Animation) (1)

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

#题目一句话要点标签🔗⭐
1 Adapting Large Language Models for Content Moderation: Pitfalls in Data Engineering and Supervised Fine-tuning 提出基于大语言模型的内容审核方法以解决数据工程挑战 large language model foundation model
2 BioBridge: Bridging Biomedical Foundation Models via Knowledge Graphs 提出BioBridge以解决生物医学单模态模型的局限性 foundation model multimodal
3 Toward a Foundation Model for Time Series Data 提出多域时间序列基础模型以提升分类性能 foundation model
4 SmoothLLM: Defending Large Language Models Against Jailbreaking Attacks 提出SmoothLLM以防御大型语言模型的越狱攻击 large language model ✅
5 UniPredict: Large Language Models are Universal Tabular Classifiers 提出UniPredict以解决表格数据预测的通用性问题 large language model
6 SteP: Stacked LLM Policies for Web Actions 提出SteP以解决大规模网络任务的动态控制问题 large language model ✅
7 Adversarial Machine Learning for Social Good: Reframing the Adversary as an Ally 提出对抗性机器学习以促进社会公益应用 large language model
8 TimeGPT-1 提出TimeGPT以解决时间序列预测问题 foundation model
9 Fine-tune Language Models to Approximate Unbiased In-context Learning 提出RICL算法以解决语言模型中的偏见问题 large language model
10 DISTFLASHATTN: Distributed Memory-efficient Attention for Long-context LLMs Training 提出DISTFLASHATTN以解决长上下文LLMs训练中的内存效率问题 large language model ✅

🔬 支柱二：RL算法与架构 (RL & Architecture) (8 篇)

#题目一句话要点标签🔗⭐
11 Small batch deep reinforcement learning 提出小批量深度强化学习以提升性能 reinforcement learning deep reinforcement learning
12 Beyond One-Preference-Fits-All Alignment: Multi-Objective Direct Preference Optimization 提出多目标直接偏好优化以解决单一模型适应性不足问题 reinforcement learning RLHF DPO ✅
13 How the level sampling process impacts zero-shot generalisation in deep reinforcement learning 提出自监督环境设计以提升深度强化学习的零-shot泛化能力 reinforcement learning deep reinforcement learning
14 PyDCM: Custom Data Center Models with Reinforcement Learning for Sustainability 提出PyDCM以优化数据中心的可持续性问题 reinforcement learning deep reinforcement learning
15 Certifiably Robust Graph Contrastive Learning 提出可证明鲁棒的图对比学习框架以应对对抗攻击 representation learning contrastive learning ✅
16 Safe Exploration in Reinforcement Learning: A Generalized Formulation and Algorithms 提出一种广义安全探索算法以解决强化学习中的安全问题 reinforcement learning
17 Constraint-Conditioned Policy Optimization for Versatile Safe Reinforcement Learning 提出条件约束策略优化以解决多样化安全强化学习问题 reinforcement learning
18 LESSON: Learning to Integrate Exploration Strategies for Reinforcement Learning via an Option Framework 提出统一框架以整合强化学习中的探索策略 reinforcement learning

🔬 支柱一：机器人控制 (Robot Control) (2 篇)

#题目一句话要点标签🔗⭐
19 Strategic Evaluation: Subjects, Evaluators, and Society 提出评估模型以解决算法决策中的道德困境 manipulation
20 Burning the Adversarial Bridges: Robust Windows Malware Detection Against Binary-level Mutations 提出图基方法以增强二进制恶意软件检测的鲁棒性 manipulation

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

#题目一句话要点标签🔗⭐
21 Targeted Adversarial Attacks on Generalizable Neural Radiance Fields 提出针对通用神经辐射场的对抗攻击方法以应对安全隐患 NeRF neural radiance field scene reconstruction

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

#题目一句话要点标签🔗⭐
22 LaTeX: Language Pattern-aware Triggering Event Detection for Adverse Experience during Pandemics 提出语言模式感知触发事件检测以应对疫情中的不良经历 PULSE

⬅️ 返回 cs.LG 首页 · 🏠 返回主页