code for "TourMLLM: A Retrieval-Augmented Multimodal Large Language Model for Multitask Learning in the Tourism Domain" in ACM ICMR2025 (Best Paper Award)
This paper presents TourMLLM, a multimodal model for tourism supporting six tasks, from landmark recognition to personalized recommendations. By using retrieval-augmented instruction tuning and preference optimization, it improves accuracy and efficiency, outperforming GPT-4o on all tasks.
[paper] [supplemental material]
Shell
60.0%
Jupyter Notebook
38.1%
Python
1.9%
code for "TourMLLM: A Retrieval-Augmented Multimodal Large Language Model for Multitask Learning in the Tourism Domain" in ACM ICMR2025 (Best Paper Award)
This paper presents TourMLLM, a multimodal model for tourism supporting six tasks, from landmark recognition to personalized recommendations. By using retrieval-augmented instruction tuning and preference optimization, it improves accuracy and efficiency, outperforming GPT-4o on all tasks.
[paper] [supplemental material]
Shell
60.0%
Jupyter Notebook
38.1%
Python
1.9%