Skip to main content
Advertisement
Browse Subject Areas
?

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

< Back to Article

Fig 1.

Spatial and temporal dependency among road sensors.

(a) Sensors 1 & 6 or sensors 4 & 5 show similar patterns, whereas they differ‌‌ from sensor 2. (b) A daily traffic pattern is observed on three consecutive days.

More »

Fig 1 Expand

Table 1.

Summary of Notation Used in the Framework.

More »

Table 1 Expand

Fig 2.

Overall architecture of the proposed Decomposition-based Dynamic Graph Adaptation of Large Language Models for Spatiotemporal Traffic Forecasting (DG-LLM) framework.

More »

Fig 2 Expand

Fig 3.

Dynamic Graph Learning Pipeline for Mode-Specific Adjacency Construction.

More »

Fig 3 Expand

Fig 4.

Mode-wise data processing through spatiotemporal embedding with graph-aware LLM backbone.

More »

Fig 4 Expand

Table 2.

Statistics of the traffic forecasting datasets used in this study.

More »

Table 2 Expand

Table 3.

Short-Term comparison on NYC-Taxi Dataset (Pick-up and Drop-off).

More »

Table 3 Expand

Table 4.

Short-Term comparison on CH-Bike Dataset (Pick-up and Drop-off).

More »

Table 4 Expand

Table 5.

Short-Term comparison on PeMS Dataset (PeMS04 and PeMS08).

More »

Table 5 Expand

Fig 5.

Statistical significance of DG-LLM improvements.

Heatmaps show percentage reduction in MAE and RMSE relative to baseline models (n = 5 seeds), along with corresponding significance levels. Colors denote significance: purple (p < 0.001), red (p < 0.01), orange (p < 0.05), and grey (not significant). Upward arrows (↑) indicate error reduction by DG-LLM.

More »

Fig 5 Expand

Fig 6.

Mean Absolute Error (MAE) comparison of DG-LLM against baseline models across 12 prediction horizons for all datasets.

Shaded areas represent the 95% confidence interval.

More »

Fig 6 Expand

Table 6.

Long-Term Comparison of NYC-Taxi Drop-off and CH-Bike Drop-off.

More »

Table 6 Expand

Table 7.

Ablation Study Results on NYC-Taxi and CH-Bike Drop-off Prediction.

More »

Table 7 Expand

Fig 7.

Relative MAE and RMSE impact of ablation studies on the full model.

More »

Fig 7 Expand

Fig 8.

MAE and RMSE comparison of curriculum learning strategy on NYC-Taxi and CH-Bike Dataset.

More »

Fig 8 Expand

Fig 9.

VMD level comparisons on the NYC-Taxi Drop-off dataset.

More »

Fig 9 Expand

Fig 10.

MAE & RMSE Comparisons for Different Unfrozen Layers on the NYC-Taxi and CH-Bike Datasets.

More »

Fig 10 Expand

Fig 11.

MAE and RMSE Comparison on Different Pretrained and Non-Pretrained Backbones.

More »

Fig 11 Expand

Fig 12.

MAE & RMSE Comparisons for Different Missing Rates on the NYC-Taxi and CH-Bike Datasets.

More »

Fig 12 Expand

Table 8.

Computational efficiency of the proposed framework across datasets. Parameter counts are in millions (M).

More »

Table 8 Expand

Fig 13.

Mode-dependent graphs learned from decomposed traffic signals.

(a) Low Frequency, (b) Mid Frequency, and (c) High Frequency.

More »

Fig 13 Expand

Fig 14.

Traffic forecasting visualization:

(a) Short-term, (b) Long-term.

More »

Fig 14 Expand

Fig 15.

Traffic forecasting results:

(a) Taxi drop-off dataset, (b) CH-Bike drop-off dataset, (c) PeMS04 dataset.

More »

Fig 15 Expand

Fig 16.

MAE and RMSE for zero-shot cross-dataset transfer performance across varying urban modalities and data scales.

More »

Fig 16 Expand