# TrajSceneLLM: A Multimodal Perspective on Semantic GPS Trajectory Analysis

arXiv: https://arxiv.org/abs/2506.16401 · HTML: https://mlsanity.com/paper/2506.16401/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 5 of its 20 references that are in the map.
- First posted on arXiv 19 Jun 2025.

## Builds on (most cited first, 5 of 5)

- [Seed1.5-VL Technical Report](https://mlsanity.com/paper/2505.07062/) (2025)
- [Large Language Models for Spatial Trajectory Patterns Mining](https://mlsanity.com/paper/2310.04942/) (2023)
- [Inferring transportation modes from GPS trajectories using a convolutional neural network](https://mlsanity.com/paper/1804.02386/) (2018)
- [Trajectory Data Management and Mining: A Survey from Deep Learning to the LLM Era](https://mlsanity.com/paper/2403.14151/) (2024)
- [RED: Effective Trajectory Representation Learning with Comprehensive Information](https://mlsanity.com/paper/2411.15096/) (2024)

## Abstract (arXiv)

GPS trajectory data reveals valuable patterns of human mobility and urban dynamics, supporting a variety of spatial applications. However, traditional methods often struggle to extract deep semantic representations and incorporate contextual map information. We propose TrajSceneLLM, a multimodal perspective for enhancing semantic understanding of GPS trajectories. The framework integrates visualized map images (encoding spatial context) and textual descriptions generated through LLM reasoning (capturing temporal sequences and movement dynamics). Separate embeddings are generated for each modality and then concatenated to produce trajectory scene embeddings with rich semantic content which are further paired with a simple MLP classifier. We validate the proposed framework on Travel Mode Identification (TMI), a critical task for analyzing travel choices and understanding mobility behavior. Our experiments show that these embeddings achieve significant performance improvement, highlighting the advantage of our LLM-driven method in capturing deep spatio-temporal dependencies and reducing reliance on handcrafted features. This semantic enhancement promises significant potential for diverse downstream applications and future research in geospatial artificial intelligence. The source code and dataset are publicly available at: https://github.com/februarysea/TrajSceneLLM.