cs.LG（2024-07-12）

📊 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) (7 🔗1) 支柱二：RL算法与架构 (RL & Architecture) (6 🔗1) 支柱四：生成式动作 (Generative Motion) (1)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (7 篇)

#题目一句话要点标签🔗⭐
1 Unifying Sequences, Structures, and Descriptions for Any-to-Any Protein Generation with the Large Multimodal Model HelixProtX HelixProtX：基于多模态大模型，实现蛋白质序列、结构和描述的任意模态生成。 large language model multimodal
2 Foundation Models for the Electric Power Grid 提出GridFM：基于图神经网络的电力系统基础模型，应对能源转型和气候变化挑战。 foundation model
3 FedsLLM: Federated Split Learning for Large Language Models over Communication Networks 提出FedsLLM框架，结合LoRA与联邦切分学习，优化无线网络中大语言模型部署。 large language model
4 Leveraging large language models for nano synthesis mechanism explanation: solid foundations or mere conjectures? 利用大语言模型解释纳米合成机制：坚实基础还是纯粹猜测？ large language model
5 GOFA: A Generative One-For-All Model for Joint Graph Language Modeling 提出GOFA：一种用于联合图语言建模的生成式One-For-All模型 large language model foundation model ✅
6 Constructing Concept-based Models to Mitigate Spurious Correlations with Minimal Human Effort 提出一种利用多模态大模型构建概念瓶颈模型的方法，以缓解虚假相关性问题。 foundation model
7 Accuracy is Not All You Need 揭示压缩LLM精度相似但行为迥异现象，提出KL散度和翻转作为评估指标 large language model

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

#题目一句话要点标签🔗⭐
8 PAIL: Performance based Adversarial Imitation Learning Engine for Carbon Neutral Optimization 提出基于性能的对抗模仿学习引擎PAIL，用于碳中和优化。 reinforcement learning deep reinforcement learning DRL
9 Aligning Diffusion Behaviors with Q-functions for Efficient Continuous Control 提出EDA算法，通过与Q函数对齐的扩散模型实现高效连续控制 reinforcement learning offline reinforcement learning diffusion policy
10 URRL-IMVC: Unified and Robust Representation Learning for Incomplete Multi-View Clustering 提出URRL-IMVC，通过统一鲁棒表示学习解决不完全多视图聚类问题 representation learning contrastive learning
11 On the Role of Discrete Tokenization in Visual Representation Learning 提出ClusterMIM，通过新型离散token化方法提升视觉表征学习的泛化能力。 representation learning contrastive learning ✅
12 Communication-Aware Reinforcement Learning for Cooperative Adaptive Cruise Control 提出通信感知强化学习(CA-RL)以提升CACC系统中MARL的可扩展性 reinforcement learning
13 HiPPO-Prophecy: State-Space Models can Provably Learn Dynamical Systems in Context HiPPO-Prophecy：提出一种新型SSM权重构造方法，实现动态系统上下文学习。 SSM state space model

🔬 支柱四：生成式动作 (Generative Motion) (1 篇)

#题目一句话要点标签🔗⭐
14 Any-Property-Conditional Molecule Generation with Self-Criticism using Spanning Trees 提出STGG+模型，利用生成树和自批判机制实现任意属性条件下的分子生成 classifier-free guidance

⬅️ 返回 cs.LG 首页 · 🏠 返回主页