Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
119 tokens/sec
GPT-4o
56 tokens/sec
Gemini 2.5 Pro Pro
43 tokens/sec
o3 Pro
6 tokens/sec
GPT-4.1 Pro
47 tokens/sec
DeepSeek R1 via Azure Pro
28 tokens/sec
2000 character limit reached

UrbanGPT: Spatio-Temporal Large Language Models (2403.00813v3)

Published 25 Feb 2024 in cs.CL, cs.AI, and cs.CY

Abstract: Spatio-temporal prediction aims to forecast and gain insights into the ever-changing dynamics of urban environments across both time and space. Its purpose is to anticipate future patterns, trends, and events in diverse facets of urban life, including transportation, population movement, and crime rates. Although numerous efforts have been dedicated to developing neural network techniques for accurate predictions on spatio-temporal data, it is important to note that many of these methods heavily depend on having sufficient labeled data to generate precise spatio-temporal representations. Unfortunately, the issue of data scarcity is pervasive in practical urban sensing scenarios. Consequently, it becomes necessary to build a spatio-temporal model with strong generalization capabilities across diverse spatio-temporal learning scenarios. Taking inspiration from the remarkable achievements of LLMs, our objective is to create a spatio-temporal LLM that can exhibit exceptional generalization capabilities across a wide range of downstream urban tasks. To achieve this objective, we present the UrbanGPT, which seamlessly integrates a spatio-temporal dependency encoder with the instruction-tuning paradigm. This integration enables LLMs to comprehend the complex inter-dependencies across time and space, facilitating more comprehensive and accurate predictions under data scarcity. To validate the effectiveness of our approach, we conduct extensive experiments on various public datasets, covering different spatio-temporal prediction tasks. The results consistently demonstrate that our UrbanGPT, with its carefully designed architecture, consistently outperforms state-of-the-art baselines. These findings highlight the potential of building LLMs for spatio-temporal learning, particularly in zero-shot scenarios where labeled data is scarce.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (51)
  1. Adaptive Graph Convolutional Recurrent Network for Traffic Forecasting. In NeurIPS. 17804–17815.
  2. Graph of Thoughts: Solving Elaborate Problems with Large Language Models. arXiv:2308.09687
  3. Language Models Are Few-Shot Learners. In NeurIPS. 1877–1901.
  4. Graphllm: Boosting graph reasoning ability of large language model. arXiv preprint arXiv:2310.05845 (2023).
  5. Towards Spatio- Temporal Aware Traffic Time Series Forecasting. In ICDE. 2900–2913.
  6. Deepmove: Predicting human mobility with attentional recurrent networks. In WWW. 1459–1468.
  7. Attention Based Spatial-Temporal Graph Convolutional Networks for Traffic Flow Forecasting. In AAAI. 922–929.
  8. Dynamic and Multi-Faceted Spatio-Temporal Deep Learning for Traffic Speed Forecasting. In KDD. 547–555.
  9. Leveraging large language models for sequential recommendation. In Recsys. 1096–1102.
  10. DeepCrime: Attentive hierarchical recurrent networks for crime prediction. In CIKM. 1423–1432.
  11. PDFormer: Propagation Delay-Aware Dynamic Long-Range Transformer for Traffic Flow Prediction. (2023), 4365–4373.
  12. Selective cross-city transfer learning for traffic prediction via source city region re-weighting. In KDD. 731–741.
  13. Diffusion convolutional recurrent neural network: data-driven traffic forecasting. In ICLR.
  14. Spatial-Temporal Hypergraph Self-Supervised Learning for Crime Prediction. In ICDE. 2984–2996.
  15. GPT-ST: Generative Pre-Training of Spatio-Temporal Graph Neural Networks. In NeurIPS.
  16. Urbanfm: Inferring fine-grained urban flows. In KDD. 3132–3142.
  17. Deep sequence learning with auxiliary information for traffic prediction. In KDD. 537–546.
  18. Spatio-Temporal Graph Few-Shot Learning with Cross-City Knowledge Transfer. In KDD. 1162–1172.
  19. A survey on deep learning for human mobility. ACM Computing Surveys (CSUR) 55, 1 (2021), 1–44.
  20. Training language models to follow instructions with human feedback. arXiv preprint arXiv:2203.021555 (2022).
  21. Urban Traffic Prediction from Spatio-Temporal Data Using Deep Meta Learning. In KDD. ACM.
  22. Representation learning with large language models for recommendation. arXiv preprint arXiv:2310.15950 (2023).
  23. Pre-training Enhanced Spatial-temporal Graph Neural Network for Multivariate Time Series Forecasting. In KDD. 1567–1577.
  24. Multitask Vision-Language Prompt Tuning. In WACV. 5656–5667.
  25. Spatial-Temporal Synchronous Graph Convolutional Networks: A New Framework for Spatial-Temporal Network Data Forecasting. In AAAI. 914–921.
  26. Graphgpt: Graph instruction tuning for large language models. arXiv preprint arXiv:2310.13023 (2023).
  27. Llama: Open and efficient foundation language models. arXiv preprint arXiv:2302.13971 (2023).
  28. Llama 2: Open foundation and fine-tuned chat models. arXiv preprint arXiv:2307.09288 (2023).
  29. GSNet: Learning Spatial-Temporal Correlations from Geographical and Semantic Aspects for Traffic Accident Risk Forecasting. AAAI (2021), 4402–4409.
  30. Pattern expansion and consolidation on evolving graphs for continual traffic prediction. In KDD. 2223–2232.
  31. Crime rate inference with big data. In KDD. 635–644.
  32. LibCity: An Open Library for Traffic Prediction. In SIGSPATIAL. 145–148.
  33. Traffic Flow Prediction via Spatial Temporal Graph Neural Network. In WWW. 1082–1092.
  34. Llmrec: Large language models with graph augmentation for recommendation. arXiv preprint arXiv:2311.00423 (2023).
  35. Connecting the Dots: Multivariate Time Series Forecasting with Graph Neural Networks. In KDD. 753–763.
  36. Graph wavenet for deep spatial-temporal graph modeling. In IJCAI.
  37. Learning from multiple cities: A meta-learning approach for spatial-temporal prediction. In WWW. 2181–2191.
  38. Deep Multi-View Spatial-Temporal Network for Taxi Demand Prediction. In AAAI.
  39. Coupled Layer-wise Graph Convolution for Transportation Demand Prediction. In AAAI. 4617–4625.
  40. ST-MVL: filling missing values in geo-sensory time series data. In IJCAI.
  41. Spatio-temporal graph convolutional networks: A deep learning framework for traffic forecasting. In IJCAI.
  42. Deep Learning: A Generic Approach for Extreme Condition Traffic Forecasting. In SDM. 777–785.
  43. GLM-130B: An Open Bilingual Pre-trained Model. In ICLR.
  44. Deep Spatio-Temporal Residual Networks for Citywide Crowd Flows Prediction. In AAAI.
  45. Automated Spatio-Temporal Graph Contrastive Learning. In WWW. 295–305.
  46. T-GCN: A Temporal Graph Convolutional Network for Traffic Prediction. Transactions on Intelligent Transportation Systems (TITS) (2020), 3848–3858.
  47. Dynamic Hypergraph Structure Learning for Traffic Flow Forecasting. ICDE.
  48. GMAN: A Graph Multi-Attention Network for Traffic Prediction. In AAAI. 1234–1241.
  49. Judging LLM-as-a-judge with MT-Bench and Chatbot Arena. arXiv:2306.05685
  50. Learning to prompt for vision-language models. International Journal of Computer Vision 130, 9 (2022), 2337–2348.
  51. MiniGPT-4: Enhancing Vision-Language Understanding with Advanced Large Language Models. arXiv:2304.10592
User Edit Pencil Streamline Icon: https://streamlinehq.com
Authors (8)
  1. Zhonghang Li (8 papers)
  2. Lianghao Xia (65 papers)
  3. Jiabin Tang (15 papers)
  4. Yong Xu (432 papers)
  5. Lei Shi (262 papers)
  6. Long Xia (25 papers)
  7. Dawei Yin (165 papers)
  8. Chao Huang (244 papers)
Citations (19)
View on Semantic Scholar

Summary

We haven't generated a summary for this paper yet.

Ai Sparkles Streamline Icon: https://streamlinehq.com Summarize Now