Salvato in:
| Autori principali: | , , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | https://arxiv.org/abs/2511.07062 |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866911293614587904 |
|---|---|
| author | Zhang, Yimei Shen, Guojiang Ning, Kaili Ren, Tongwei Qiu, Xuebo Wang, Mengmeng Kong, Xiangjie |
| author_facet | Zhang, Yimei Shen, Guojiang Ning, Kaili Ren, Tongwei Qiu, Xuebo Wang, Mengmeng Kong, Xiangjie |
| contents | Region representation learning plays a pivotal role in urban computing by extracting meaningful features from unlabeled urban data. Analogous to how perceived facial age reflects an individual's health, the visual appearance of a city serves as its "portrait", encapsulating latent socio-economic and environmental characteristics. Recent studies have explored leveraging Large Language Models (LLMs) to incorporate textual knowledge into imagery-based urban region representation learning. However, two major challenges remain: i) difficulty in aligning fine-grained visual features with long captions, and ii) suboptimal knowledge incorporation due to noise in LLM-generated captions. To address these issues, we propose a novel pre-training framework called UrbanLN that improves Urban region representation learning through Long-text awareness and Noise suppression. Specifically, we introduce an information-preserved stretching interpolation strategy that aligns long captions with fine-grained visual semantics in complex urban scenes. To effectively mine knowledge from LLM-generated captions and filter out noise, we propose a dual-level optimization strategy. At the data level, a multi-model collaboration pipeline automatically generates diverse and reliable captions without human intervention. At the model level, we employ a momentum-based self-distillation mechanism to generate stable pseudo-targets, facilitating robust cross-modal learning under noisy conditions. Extensive experiments across four real-world cities and various downstream tasks demonstrate the superior performance of our UrbanLN. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2511_07062 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Improving Region Representation Learning from Urban Imagery with Noisy Long-Caption Supervision Zhang, Yimei Shen, Guojiang Ning, Kaili Ren, Tongwei Qiu, Xuebo Wang, Mengmeng Kong, Xiangjie Artificial Intelligence Region representation learning plays a pivotal role in urban computing by extracting meaningful features from unlabeled urban data. Analogous to how perceived facial age reflects an individual's health, the visual appearance of a city serves as its "portrait", encapsulating latent socio-economic and environmental characteristics. Recent studies have explored leveraging Large Language Models (LLMs) to incorporate textual knowledge into imagery-based urban region representation learning. However, two major challenges remain: i) difficulty in aligning fine-grained visual features with long captions, and ii) suboptimal knowledge incorporation due to noise in LLM-generated captions. To address these issues, we propose a novel pre-training framework called UrbanLN that improves Urban region representation learning through Long-text awareness and Noise suppression. Specifically, we introduce an information-preserved stretching interpolation strategy that aligns long captions with fine-grained visual semantics in complex urban scenes. To effectively mine knowledge from LLM-generated captions and filter out noise, we propose a dual-level optimization strategy. At the data level, a multi-model collaboration pipeline automatically generates diverse and reliable captions without human intervention. At the model level, we employ a momentum-based self-distillation mechanism to generate stable pseudo-targets, facilitating robust cross-modal learning under noisy conditions. Extensive experiments across four real-world cities and various downstream tasks demonstrate the superior performance of our UrbanLN. |
| title | Improving Region Representation Learning from Urban Imagery with Noisy Long-Caption Supervision |
| topic | Artificial Intelligence |
| url | https://arxiv.org/abs/2511.07062 |