Simplifying Multimodality: Unimodal Approach to Multimodal Challenges in Radiology with General-Domain Large Language Model
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2024
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866914782759616512 |
|---|---|
| author | Cho, Seonhee Kim, Choonghan Lee, Jiho Chilkunda, Chetan Choi, Sujin Yoon, Joo Heung |
| author_facet | Cho, Seonhee Kim, Choonghan Lee, Jiho Chilkunda, Chetan Choi, Sujin Yoon, Joo Heung |
| contents | Recent advancements in Large Multimodal Models (LMMs) have attracted interest in their generalization capability with only a few samples in the prompt. This progress is particularly relevant to the medical domain, where the quality and sensitivity of data pose unique challenges for model training and application. However, the dependency on high-quality data for effective in-context learning raises questions about the feasibility of these models when encountering with the inevitable variations and errors inherent in real-world medical data. In this paper, we introduce MID-M, a novel framework that leverages the in-context learning capabilities of a general-domain Large Language Model (LLM) to process multimodal data via image descriptions. MID-M achieves a comparable or superior performance to task-specific fine-tuned LMMs and other general-domain ones, without the extensive domain-specific training or pre-training on multimodal data, with significantly fewer parameters. This highlights the potential of leveraging general-domain LLMs for domain-specific tasks and offers a sustainable and cost-effective alternative to traditional LMM developments. Moreover, the robustness of MID-M against data quality issues demonstrates its practical utility in real-world medical domain applications. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2405_01591 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Simplifying Multimodality: Unimodal Approach to Multimodal Challenges in Radiology with General-Domain Large Language Model Cho, Seonhee Kim, Choonghan Lee, Jiho Chilkunda, Chetan Choi, Sujin Yoon, Joo Heung Computation and Language Artificial Intelligence Image and Video Processing Recent advancements in Large Multimodal Models (LMMs) have attracted interest in their generalization capability with only a few samples in the prompt. This progress is particularly relevant to the medical domain, where the quality and sensitivity of data pose unique challenges for model training and application. However, the dependency on high-quality data for effective in-context learning raises questions about the feasibility of these models when encountering with the inevitable variations and errors inherent in real-world medical data. In this paper, we introduce MID-M, a novel framework that leverages the in-context learning capabilities of a general-domain Large Language Model (LLM) to process multimodal data via image descriptions. MID-M achieves a comparable or superior performance to task-specific fine-tuned LMMs and other general-domain ones, without the extensive domain-specific training or pre-training on multimodal data, with significantly fewer parameters. This highlights the potential of leveraging general-domain LLMs for domain-specific tasks and offers a sustainable and cost-effective alternative to traditional LMM developments. Moreover, the robustness of MID-M against data quality issues demonstrates its practical utility in real-world medical domain applications. |
| title | Simplifying Multimodality: Unimodal Approach to Multimodal Challenges in Radiology with General-Domain Large Language Model |
| topic | Computation and Language Artificial Intelligence Image and Video Processing |
| url | https://arxiv.org/abs/2405.01591 |