Cross-domain Traffic Prediction on Milan-SMS
7.2017MAEST-Vision-LLM
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| ST-Vision-LLMInput sequence length (S)=12, Prediction horizon (K)=362025.10 | 7.2017 | 17.6204 | 0.7213 | |
| GATGPTInput sequence length (S)=12, Prediction horizon (K)=362025.10 | 8.9146 | 18.5483 | 0.7593 | |
| ST-LLMInput sequence length (S)=12, Prediction horizon (K)=362025.10 | 8.9303 | 18.9329 | 0.775 | |
| GCNGPTInput sequence length (S)=12, Prediction horizon (K)=362025.10 | 9.592 | 18.8464 | 0.7715 | |
| MCSTGCNInput sequence length (S)=12, Prediction horizon (K)=362025.10 | 15.432 | 26.2068 | 1.0728 | |
| ARIMAInput sequence length (S)=12, Prediction horizon (K)=362025.10 | 15.7524 | 26.3374 | 1.0781 | |
| STNInput sequence length (S)=12, Prediction horizon (K)=362025.10 | 16.256 | 27.39 | 1.1212 | |
| Time-LLMInput sequence length (S)=12, Prediction horizon (K)=362025.10 | 16.8151 | 29.2006 | 1.1953 | |
| ACFMInput sequence length (S)=12, Prediction horizon (K)=362025.10 | 24.3491 | 37.5989 | 1.5391 |