HotelMatch-LLM: Joint Multi-Task Training of Small and Large Language Models for Efficient Multimodal Hotel Retrieval
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2025
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866910995342950400 |
|---|---|
| author | Askari, Arian Stergiadis, Emmanouil Gusev, Ilya Beladev, Moran |
| author_facet | Askari, Arian Stergiadis, Emmanouil Gusev, Ilya Beladev, Moran |
| contents | We present HotelMatch-LLM, a multimodal dense retrieval model for the travel domain that enables natural language property search, addressing the limitations of traditional travel search engines which require users to start with a destination and editing search parameters. HotelMatch-LLM features three key innovations: (1) Domain-specific multi-task optimization with three novel retrieval, visual, and language modeling objectives; (2) Asymmetrical dense retrieval architecture combining a small language model (SLM) for efficient online query processing and a large language model (LLM) for embedding hotel data; and (3) Extensive image processing to handle all property image galleries. Experiments on four diverse test sets show HotelMatch-LLM significantly outperforms state-of-the-art models, including VISTA and MARVEL. Specifically, on the test set -- main query type -- we achieve 0.681 for HotelMatch-LLM compared to 0.603 for the most effective baseline, MARVEL. Our analysis highlights the impact of our multi-task optimization, the generalizability of HotelMatch-LLM across LLM architectures, and its scalability for processing large image galleries. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2506_07296 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | HotelMatch-LLM: Joint Multi-Task Training of Small and Large Language Models for Efficient Multimodal Hotel Retrieval Askari, Arian Stergiadis, Emmanouil Gusev, Ilya Beladev, Moran Information Retrieval Artificial Intelligence Computer Vision and Pattern Recognition We present HotelMatch-LLM, a multimodal dense retrieval model for the travel domain that enables natural language property search, addressing the limitations of traditional travel search engines which require users to start with a destination and editing search parameters. HotelMatch-LLM features three key innovations: (1) Domain-specific multi-task optimization with three novel retrieval, visual, and language modeling objectives; (2) Asymmetrical dense retrieval architecture combining a small language model (SLM) for efficient online query processing and a large language model (LLM) for embedding hotel data; and (3) Extensive image processing to handle all property image galleries. Experiments on four diverse test sets show HotelMatch-LLM significantly outperforms state-of-the-art models, including VISTA and MARVEL. Specifically, on the test set -- main query type -- we achieve 0.681 for HotelMatch-LLM compared to 0.603 for the most effective baseline, MARVEL. Our analysis highlights the impact of our multi-task optimization, the generalizability of HotelMatch-LLM across LLM architectures, and its scalability for processing large image galleries. |
| title | HotelMatch-LLM: Joint Multi-Task Training of Small and Large Language Models for Efficient Multimodal Hotel Retrieval |
| topic | Information Retrieval Artificial Intelligence Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2506.07296 |