Abstract
Large-scale vectorized HD maps provide structured road information that is essential for perception, localization, and planning in autonomous driving. Constructing such maps requires aggregating noisy, fragmented, and overlapping local predictions collected along a vehicle trajectory into a coherent global map. Existing aggregation methods typically rely on hand-crafted rules for fragment association and refinement. However, a fixed set of thresholds cannot effectively handle variations in road structures and prediction errors, often requiring detector-specific tuning or manual adjustment. To address this limitation, we propose MapMergeLLM, a data-driven framework that formulates vectorized map aggregation as conditional sequence generation with a large language model. Given serialized local vectorized maps, our model directly predicts the aggregated global map polylines. To reduce dependence on any particular upstream detector, we train the model on synthetic local maps generated from clean vector maps using corruptions that simulate representative prediction errors. We further introduce a coordinate tokenizer with geometry-aware pretraining to precisely represent map coordinates. In addition, we propose a line-level association loss that explicitly supervises correspondences between local observations of the same map element. Experiments on Argoverse2 and nuScenes using multiple recent upstream detectors demonstrate that MapMergeLLM substantially outperforms heuristic and optimization-based aggregation baselines without detector-specific retraining.
Keywords
Subject
Publication details
- Journal
- Not available
- Open access
- Green open access
Cite this article
APA 7
Li, Z., Chen, Y. T., Wang, X., He, W., Shen, H. W., & Ren, L. (2026). From Fragments to Global Maps: Learning Vectorized Map Aggregation with Large Language Models. https://omanscience.com/en/articles/from-fragments-to-global-maps-learning-vectorized-map-aggregation-with-large-language-models
MLA 9
Li, Ziwei, et al. "From Fragments to Global Maps: Learning Vectorized Map Aggregation with Large Language Models." https://omanscience.com/en/articles/from-fragments-to-global-maps-learning-vectorized-map-aggregation-with-large-language-models.
Chicago (author–date)
Li, Ziwei, Yi-Tang Chen, Xiaoqi Wang, Wenbin He, Han-Wei Shen, and Liu Ren. 2026. "From Fragments to Global Maps: Learning Vectorized Map Aggregation with Large Language Models." https://omanscience.com/en/articles/from-fragments-to-global-maps-learning-vectorized-map-aggregation-with-large-language-models.
Harvard
Li, Z., Chen, Y. T., Wang, X., He, W., Shen, H. W. and Ren, L. (2026) 'From Fragments to Global Maps: Learning Vectorized Map Aggregation with Large Language Models', Available at: https://omanscience.com/en/articles/from-fragments-to-global-maps-learning-vectorized-map-aggregation-with-large-language-models.
Vancouver
Li Z, Chen YT, Wang X, He W, Shen HW, Ren L. From Fragments to Global Maps: Learning Vectorized Map Aggregation with Large Language Models. https://omanscience.com/en/articles/from-fragments-to-global-maps-learning-vectorized-map-aggregation-with-large-language-models
IEEE
Z. Li, Y. T. Chen, X. Wang, W. He, H. W. Shen, and L. Ren, "From Fragments to Global Maps: Learning Vectorized Map Aggregation with Large Language Models," https://omanscience.com/en/articles/from-fragments-to-global-maps-learning-vectorized-map-aggregation-with-large-language-models.