NeurIPS 2024 | 时空数据（Spatial-Temporal）论文总结

NeurIPS 2024于2024年12月10号-12月15号在加拿大温哥华举行（Vancouver, Canada）,录取率25.8% 本文总结了NeurIPS 2024有关时空数据（spatial-temporal data）的相关论文，主要包含气象预测，时空插补，轨迹生成，交通模拟，信控优化，异常检测以及LLM在时空数据的应用等内容，如有疏漏，欢迎大家补充。 时空数据Topic：气象预测，时空插补，轨迹生成，交通模拟，信控优化，异常检测以及LLM在时空数据的应用等内容。总计20篇，其中正会12篇，D&B Track8篇 点击文末阅读原文跳转笔者知乎链接（跳转论文链接更方便）。

1. [Oral] Get Rid of Isolation: A Continuous Multi-task Spatio-Temporal Learning Framework
2. [Spoatlight] DiffLight: A Partial Rewards Conditioned Diffusion Model for Traffic Signal Control with Missing Data
3. Mobility-LLM: Learning Visiting Intentions and Travel Preference from Human Mobility Data with Large Language Models
4. Addressing Spatial-Temporal Heterogeneity: General Mixed Time Series Analysis via Latent Continuity Recovery and Alignment
5. Causal Deciphering and Inpainting in Spatio-Temporal Dynamics via Diffusion Model
6. Generating Origin-Destination Matrices in Neural Spatial Interaction Models
7. Language-Driven Interactive Traffic Trajectory Generation
8. NetworkGym: Reinforcement Learning Environments for Multi-Access Traffic Management in Network Simulation
9. Sample-efficient Simulation-based Inference for Urban Travel Demand Calibration
10. Large Language Models as Urban Residents: An LLM Agent Framework for Personal Mobility Generation
11. Learning from Highly Sparse Spatio-temporal Data
12. UrbanKGent: A Unified Large Language Model Agent Framework for Urban Knowledge Graph Construction

D&B Track

1. LogiCity: Advancing Neuro-Symbolic AI with Abstract Urban Simulation
2. TorchSpatial: A Location Encoding Framework and Benchmark for Spatial Representation Learning
3. FT-AED: Benchmark Dataset for Early Freeway Traffic Anomalous Event Detection
4. Terra: A Multimodal Spatio-Temporal Dataset Spanning the Earth
5. UrbanDataLayer: A Unified Data Pipeline for Urban Science
6. [Oral] ChaosBench: A Multi-Channel, Physics-Based Benchmark for Subseasonal-to-Seasonal Climate Prediction
7. SolarCube: An Integrative Benchmark Dataset Harnessing Satellite and In-situ Observations for Large-scale Solar Energy Forecasting
8. [Spotlight] GeoLife: Spacial Plant Species Prediction Dataset

1 [Oral] Get Rid of Isolation: A Continuous Multi-task Spatio-Temporal Learning Framework

链接：https://nips.cc/virtual/2024/poster/93311

作者：Zhongchao Yi, Zhengyang Zhou, Qihe Huang, Yanjiang Chen, Liheng Yu, Xu Wang, Yang Wang

关键词：多任务，持续学习

2 [Spoatlight] DiffLight: A Partial Rewards Conditioned Diffusion Model for Traffic Signal Control with Missing Data

链接：https://nips.cc/virtual/2024/poster/96278

作者：Hanyang Chen, Yang Jiang, Shengnan Guo, Xiaowei Mao, Youfang Lin, Huaiyu Wan

关键词：信控优化，扩散模型

DiffLight

3 Mobility-LLM: Learning Visiting Intentions and Travel Preference from Human Mobility Data with Large Language Models

链接：https://neurips.cc/virtual/2024/poster/96914

作者：Letian Gong · Yan Lin · Xinyue Zhang · Yiwen Lu · Xuedi Han · Yichen Liu · Shengnan Guo · Youfang Lin · Huaiyu Wan

关键词：基于位置服务，LLM

Mobility-LLM

2-3的模型图和信息来源于：实验室8篇学术论文被人工智能顶会NeurlPS 2024录用

4 Addressing Spatial-Temporal Heterogeneity: General Mixed Time Series Analysis via Latent Continuity Recovery and Alignment

链接：https://nips.cc/virtual/2024/poster/96039

作者：Jiawei Chen, 春晖 赵

关键词：时空异质性，对齐，恢复

5 Causal Deciphering and Inpainting in Spatio-Temporal Dynamics via Diffusion Model

链接：https://nips.cc/virtual/2024/poster/96868

作者：Yifan Duan, Jian Zhao, pengcheng, Junyuan Mao, Hao Wu, Jingyu Xu, shilong wang, Caoyuan Ma, Kai Wang, Kun Wang, Xuelong Li

关键词：因果，扩散模型

6 Generating Origin-Destination Matrices in Neural Spatial Interaction Models

链接：https://nips.cc/virtual/2024/poster/94266

作者：Ioannis Zachos, Mark Girolami, Theodoros Damoulas

关键词：OD矩阵生成

7 Language-Driven Interactive Traffic Trajectory Generation

链接：https://nips.cc/virtual/2024/poster/96845

arXiv: https://arxiv.org/abs/2405.15388

代码：https://github.com/X1a-jk/InteractTraj.git

作者：Junkai XIA, Chenxin Xu, Qingyao Xu, Yanfeng Wang, Siheng Chen

关键词：轨迹生成，语言模型

InteractTraj

8 NetworkGym: Reinforcement Learning Environments for Multi-Access Traffic Management in Network Simulation

链接：https://nips.cc/virtual/2024/poster/97863

作者：Momin Haider, Ming Yin, Menglei Zhang, Arpit Gupta, Jing Zhu, Yu-Xiang Wang

关键词：交通模拟，强化学习

9 Sample-efficient Simulation-based Inference for Urban Travel Demand Calibration

链接：https://nips.cc/virtual/2024/poster/94187

作者：Sam Griesemer, Defu Cao, Zijun Cui, Carolina Osorio, Yan Liu

关键词：城市交通需求校准推理

10 Large Language Models as Urban Residents: An LLM Agent Framework for Personal Mobility Generation

链接：https://nips.cc/virtual/2024/poster/96855

arXiv: https://arxiv.org/abs/2402.14744

作者：WANG JIAWEI, Renhe Jiang, Chuang Yang, Zengqing Wu, makoto onizuka, Ryosuke Shibasaki, Noboru Koshizuka, Chuan Xiao

关键词：轨迹生成，LLM

AI论文速读 | 大语言模型作为城市居民——利用LLM智能体框架生成人类移动轨迹

LLMob

11 Learning from Highly Sparse Spatio-temporal Data

链接：https://nips.cc/virtual/2024/poster/93436

作者：Leyan Deng, Chenwang Wu, Defu Lian, Enhong Chen

关键词：稀疏性，插补

12 UrbanKGent: A Unified Large Language Model Agent Framework for Urban Knowledge Graph Construction

链接：https://nips.cc/virtual/2024/poster/95400

arXiv: https://arxiv.org/abs/2402.06861

代码：https://github.com/usail-hkust/UrbanKGent

作者：Yansong Ning, Hao Liu

关键词：城市知识图谱构建

UrbanKGent

D&B Track

13 LogiCity: Advancing Neuro-Symbolic AI with Abstract Urban Simulation

链接：https://nips.cc/virtual/2024/poster/97738

Demo：https://jaraxxus-me.github.io/LogiCity/

作者：Bowen Li, Zhaoyu Li, Qiwei Du, Jinqi Luo, Wenshan Wang, Yaqi Xie, Simon Stepputtis, Chen Wang, Katia Sycara, Pradeep Ravikumar, Alexander Gray, Xujie Si, Sebastian Scherer

关键词：城市模拟，神经符号

14 TorchSpatial: A Location Encoding Framework and Benchmark for Spatial Representation Learning

链接：https://nips.cc/virtual/2024/poster/97807

arXiv: https://arxiv.org/abs/2406.15658

代码：https://github.com/seai-lab/TorchSpatial

作者：Nemin Wu · Qian Cao · Zhangyu Wang · Zeping Liu · Yanlin Qi · Jielu Zhang · Joshua Ni · X. Yao · Hongxu Ma · Lan Mu · Stefano Ermon · Tanuja Ganu · Akshay Nambi · Ni Lao · Gengchen Mai

关键词：空间表示学习

TorchSpatial Overall Framework

15 FT-AED: Benchmark Dataset for Early Freeway Traffic Anomalous Event Detection

链接：https://nips.cc/virtual/2024/poster/97562

arXiv: https://arxiv.org/abs/2406.15283

代码：https://acoursey3.github.io/ft-aed/

作者：Austin Coursey · Junyi Ji · Marcos Quinones Grueiro · William Barbour · Yuhang Zhang · Tyler Derr · Gautam Biswas · Daniel Work

关键词：异常检测，数据集

FT-AED

16 Terra: A Multimodal Spatio-Temporal Dataset Spanning the Earth

链接：https://nips.cc/virtual/2024/poster/97768

作者：Wei Chen · Xixuan Hao · Yuankai Wu · Yuxuan Liang

关键词：地球气象数据集，多模态

17 UrbanDataLayer: A Unified Data Pipeline for Urban Science

链接：https://nips.cc/virtual/2024/poster/97837

作者：Yiheng Wang · Tianyu Wang · YuYing Zhang · Hongji Zhang · Haoyu Zheng · Guanjie Zheng · Linghe Kong

关键词：流水线，城市科学

18 [Oral] ChaosBench: A Multi-Channel, Physics-Based Benchmark for Subseasonal-to-Seasonal Climate Prediction

链接：https://neurips.cc/virtual/2024/oral/98017

arXiv: https://arxiv.org/abs/2402.00712

Demo：https://leap-stc.github.io/ChaosBench/

作者：Juan Nathaniel · Yongquan Qu · Tung Nguyen · Sungduk Yu · Julius Busecke · Aditya Grover · Pierre Gentine

关键词：气候预测，Benchmark

ChaosBench

19 SolarCube: An Integrative Benchmark Dataset Harnessing Satellite and In-situ Observations for Large-scale Solar Energy Forecasting

链接：https://neurips.cc/virtual/2024/poster/97653

代码： https://github.com/Ruohan-Li/SolarCube

作者：Ruohan Li · Yiqun Xie · Xiaowei Jia · Dongdong Wang · Yanhua Li · Yingxue Zhang · Zhihao Wang · Zhili Li

关键词：太阳能预测，benchmark

SolarCube

20 [Spotlight] GeoLife: Spacial Plant Species Prediction Dataset

链接：https://arxiv.org/abs/2308.05121

Kaggle： https://www.kaggle.com/competitions/geolifeclef-2023-lifeclef-2023-x-fgvc10/

作者：Lukas Picek · Christophe Botella · Maximilien Servajean · César Leblanc · Rémi Palard · Theo Larcher · Benjamin Deneu · Diego Marcos · Pierre Bonnet · alexis joly

关键词：生态分布预测，物种多样性

GeoLifeCLEF

相关链接

NeurIPS 24 Accepted Papers：https://neurips.cc/virtual/2024/papers.html

欢迎各位作者投稿近期有关时空数据和时间序列录用的顶级会议和期刊的优秀文章解读，我们将竭诚为您宣传,共同学习进步。如有意愿,请通过后台私信与我们联系。

如果觉得有帮助还请分享，在看，点赞