Leveraging Taxonomy and LLMs for Improved Multimodal Hierarchical Classification
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866915099472560128 |
|---|---|
| author | Chen, Shijing Bouadjenek, Mohamed Reda Jameel, Shoaib Naseem, Usman Suleiman, Basem Salim, Flora D. Hacid, Hakim Razzak, Imran |
| author_facet | Chen, Shijing Bouadjenek, Mohamed Reda Jameel, Shoaib Naseem, Usman Suleiman, Basem Salim, Flora D. Hacid, Hakim Razzak, Imran |
| contents | Multi-level Hierarchical Classification (MLHC) tackles the challenge of categorizing items within a complex, multi-layered class structure. However, traditional MLHC classifiers often rely on a backbone model with independent output layers, which tend to ignore the hierarchical relationships between classes. This oversight can lead to inconsistent predictions that violate the underlying taxonomy. Leveraging Large Language Models (LLMs), we propose a novel taxonomy-embedded transitional LLM-agnostic framework for multimodality classification. The cornerstone of this advancement is the ability of models to enforce consistency across hierarchical levels. Our evaluations on the MEP-3M dataset - a multi-modal e-commerce product dataset with various hierarchical levels - demonstrated a significant performance improvement compared to conventional LLM structures. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2501_06827 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Leveraging Taxonomy and LLMs for Improved Multimodal Hierarchical Classification Chen, Shijing Bouadjenek, Mohamed Reda Jameel, Shoaib Naseem, Usman Suleiman, Basem Salim, Flora D. Hacid, Hakim Razzak, Imran Artificial Intelligence Multi-level Hierarchical Classification (MLHC) tackles the challenge of categorizing items within a complex, multi-layered class structure. However, traditional MLHC classifiers often rely on a backbone model with independent output layers, which tend to ignore the hierarchical relationships between classes. This oversight can lead to inconsistent predictions that violate the underlying taxonomy. Leveraging Large Language Models (LLMs), we propose a novel taxonomy-embedded transitional LLM-agnostic framework for multimodality classification. The cornerstone of this advancement is the ability of models to enforce consistency across hierarchical levels. Our evaluations on the MEP-3M dataset - a multi-modal e-commerce product dataset with various hierarchical levels - demonstrated a significant performance improvement compared to conventional LLM structures. |
| title | Leveraging Taxonomy and LLMs for Improved Multimodal Hierarchical Classification |
| topic | Artificial Intelligence |
| url | https://arxiv.org/abs/2501.06827 |