cs.LG（2024-10-05）

📊 共 18 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) (10 🔗1) 支柱二：RL算法与架构 (RL & Architecture) (6 🔗1) 支柱八：物理动画 (Physics-based Animation) (1) 支柱三：空间感知与语义 (Perception & Semantics) (1)

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

#题目一句话要点标签🔗⭐
1 Multimodal Large Language Models for Inverse Molecular Design with Retrosynthetic Planning Llamole：用于逆分子设计与逆合成规划的多模态大语言模型 large language model multimodal
2 An Electrocardiogram Foundation Model Built on over 10 Million Recordings with External Evaluation across Multiple Domains 提出ECGFounder心电图基础模型，基于千万级数据训练，提升多领域心电分析性能。 foundation model ✅
3 PalmBench: A Comprehensive Benchmark of Compressed Large Language Models on Mobile Platforms PalmBench：移动平台压缩大语言模型综合评测基准 large language model
4 Hyperbolic Fine-tuning for Large Language Models 提出HypLoRA，一种在双曲空间高效微调大型语言模型的方法，显著提升复杂推理任务性能。 large language model
5 Domain-Oriented Time Series Inference Agents for Reasoning and Automated Analysis 提出面向领域的时间序列推理Agent TS-Reasoner，用于复杂推理和自动化分析。 large language model foundation model
6 Improving LLM Reasoning through Scaling Inference Computation with Collaborative Verification 提出协同验证方法，通过扩展推理计算提升LLM在复杂推理任务中的性能 large language model chain-of-thought
7 Enhancing Carbon Emission Reduction Strategies using OCO and ICOS data 融合OCO和ICOS数据，提出高分辨率CO2估算方法，助力碳减排策略 multimodal
8 Black Boxes and Looking Glasses: Multilevel Symmetries, Reflection Planes, and Convex Optimization in Deep Networks 将绝对值激活的深度网络训练转化为凸Lasso问题，揭示网络对称性 large language model
9 Functional Homotopy: Smoothing Discrete Optimization via Continuous Parameters for LLM Jailbreak Attacks 提出函数同伦方法，通过连续参数平滑离散优化，提升LLM越狱攻击成功率。 large language model
10 Beyond Language: Applying MLX Transformers to Engineering Physics 提出基于MLX Transformer的物理信息模型，高效求解2D热传导问题。 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🔗⭐
11 Comprehensive Monitoring of Air Pollution Hotspots Using Sparse Sensor Networks 结合预测模型与机理方法，利用稀疏传感器网络全面监测空气污染热点 predictive model sparse sensors
12 Fundamental Limitations on Subquadratic Alternatives to Transformers 证明Transformer的子二次替代方案在文档相似性任务中存在根本性局限 Mamba state space model large language model
13 On the Global Optimality of Policy Gradient Methods in General Utility Reinforcement Learning 针对通用效用强化学习，证明策略梯度方法具有全局最优性 reinforcement learning imitation learning
14 Parametric Taylor series based latent dynamics identification neural networks 提出基于参数化泰勒级数的潜变量动力学辨识神经网络(P-TLDINets)，用于高效求解参数化偏微分方程。 latent dynamics
15 Text2Chart31: Instruction Tuning for Chart Generation with Automatic Feedback 提出Text2Chart31数据集和基于强化学习的指令调优方法，提升LLM在图表生成任务中的性能。 reinforcement learning large language model ✅
16 Reward Learning From Preference With Ties 提出BTT模型，解决RLHF中奖励学习忽略并列偏好导致的偏差问题 reinforcement learning RLHF

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

#题目一句话要点标签🔗⭐
17 From Incomplete Coarse-Grained to Complete Fine-Grained: A Two-Stage Framework for Spatiotemporal Data Reconstruction 提出DiffRecon框架，解决时空数据从粗粒度到细粒度的重建问题 spatiotemporal

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

#题目一句话要点标签🔗⭐
18 Equivariant Polynomial Functional Networks 提出MAGEP-NFN，一种高效且具有表达力的置换和尺度等变函数网络。 implicit representation

⬅️ 返回 cs.LG 首页 · 🏠 返回主页