MobEvolve: An Agentic Self-Evolving Heuristic System for Interpretable Human Mobility Generation
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2026
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866910279524155392 |
|---|---|
| author | He, Junlin Tang, Yihong Nie, Tong Qu, Ao Liang, Yuebing Alizadeh, Hamzeh Liu, Bang Ma, Wei Sun, Lijun |
| author_facet | He, Junlin Tang, Yihong Nie, Tong Qu, Ao Liang, Yuebing Alizadeh, Hamzeh Liu, Bang Ma, Wei Sun, Lijun |
| contents | Human mobility generation aims to synthesize realistic trip chains for target populations based on individual features. Existing paradigms, including deep generative models, LLM-based methods, and traditional heuristics, struggle to satisfy the complex demands of this task while simultaneously maintaining interpretability, behavioral plausibility, population-level distributional alignment, and inference efficiency. To bridge this gap, we introduce MobEvolve, the first agentic self-evolving heuristic framework for human mobility generation. MobEvolve initializes a behavior-inspired heuristic system and employs an LLM agent to iteratively evolve its internal logic. By diagnosing empirical misalignments and failure cases on a validation set, the agent proposes targeted updates and accumulates evolution memory for cumulative self-improvement. Extensive evaluations on the Singapore and Montreal benchmarks demonstrate that MobEvolve significantly outperforms state-of-the-art deep generative and LLM-based methods in individual trajectory fidelity, population-level distribution alignment, and behavioral plausibility, while preserving interpretability and high inference efficiency. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2606_01640 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | MobEvolve: An Agentic Self-Evolving Heuristic System for Interpretable Human Mobility Generation He, Junlin Tang, Yihong Nie, Tong Qu, Ao Liang, Yuebing Alizadeh, Hamzeh Liu, Bang Ma, Wei Sun, Lijun Artificial Intelligence Computation and Language Human mobility generation aims to synthesize realistic trip chains for target populations based on individual features. Existing paradigms, including deep generative models, LLM-based methods, and traditional heuristics, struggle to satisfy the complex demands of this task while simultaneously maintaining interpretability, behavioral plausibility, population-level distributional alignment, and inference efficiency. To bridge this gap, we introduce MobEvolve, the first agentic self-evolving heuristic framework for human mobility generation. MobEvolve initializes a behavior-inspired heuristic system and employs an LLM agent to iteratively evolve its internal logic. By diagnosing empirical misalignments and failure cases on a validation set, the agent proposes targeted updates and accumulates evolution memory for cumulative self-improvement. Extensive evaluations on the Singapore and Montreal benchmarks demonstrate that MobEvolve significantly outperforms state-of-the-art deep generative and LLM-based methods in individual trajectory fidelity, population-level distribution alignment, and behavioral plausibility, while preserving interpretability and high inference efficiency. |
| title | MobEvolve: An Agentic Self-Evolving Heuristic System for Interpretable Human Mobility Generation |
| topic | Artificial Intelligence Computation and Language |
| url | https://arxiv.org/abs/2606.01640 |