Explainable and Interpretable Multimodal Large Language Models: A Comprehensive Survey
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , , , , , , , , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2024
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866915045633425408 |
|---|---|
| author | Dang, Yunkai Huang, Kaichen Huo, Jiahao Yan, Yibo Huang, Sirui Liu, Dongrui Gao, Mengxi Zhang, Jie Qian, Chen Wang, Kun Liu, Yong Shao, Jing Xiong, Hui Hu, Xuming |
| author_facet | Dang, Yunkai Huang, Kaichen Huo, Jiahao Yan, Yibo Huang, Sirui Liu, Dongrui Gao, Mengxi Zhang, Jie Qian, Chen Wang, Kun Liu, Yong Shao, Jing Xiong, Hui Hu, Xuming |
| contents | The rapid development of Artificial Intelligence (AI) has revolutionized numerous fields, with large language models (LLMs) and computer vision (CV) systems driving advancements in natural language understanding and visual processing, respectively. The convergence of these technologies has catalyzed the rise of multimodal AI, enabling richer, cross-modal understanding that spans text, vision, audio, and video modalities. Multimodal large language models (MLLMs), in particular, have emerged as a powerful framework, demonstrating impressive capabilities in tasks like image-text generation, visual question answering, and cross-modal retrieval. Despite these advancements, the complexity and scale of MLLMs introduce significant challenges in interpretability and explainability, essential for establishing transparency, trustworthiness, and reliability in high-stakes applications. This paper provides a comprehensive survey on the interpretability and explainability of MLLMs, proposing a novel framework that categorizes existing research across three perspectives: (I) Data, (II) Model, (III) Training \& Inference. We systematically analyze interpretability from token-level to embedding-level representations, assess approaches related to both architecture analysis and design, and explore training and inference strategies that enhance transparency. By comparing various methodologies, we identify their strengths and limitations and propose future research directions to address unresolved challenges in multimodal explainability. This survey offers a foundational resource for advancing interpretability and transparency in MLLMs, guiding researchers and practitioners toward developing more accountable and robust multimodal AI systems. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2412_02104 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Explainable and Interpretable Multimodal Large Language Models: A Comprehensive Survey Dang, Yunkai Huang, Kaichen Huo, Jiahao Yan, Yibo Huang, Sirui Liu, Dongrui Gao, Mengxi Zhang, Jie Qian, Chen Wang, Kun Liu, Yong Shao, Jing Xiong, Hui Hu, Xuming Computation and Language The rapid development of Artificial Intelligence (AI) has revolutionized numerous fields, with large language models (LLMs) and computer vision (CV) systems driving advancements in natural language understanding and visual processing, respectively. The convergence of these technologies has catalyzed the rise of multimodal AI, enabling richer, cross-modal understanding that spans text, vision, audio, and video modalities. Multimodal large language models (MLLMs), in particular, have emerged as a powerful framework, demonstrating impressive capabilities in tasks like image-text generation, visual question answering, and cross-modal retrieval. Despite these advancements, the complexity and scale of MLLMs introduce significant challenges in interpretability and explainability, essential for establishing transparency, trustworthiness, and reliability in high-stakes applications. This paper provides a comprehensive survey on the interpretability and explainability of MLLMs, proposing a novel framework that categorizes existing research across three perspectives: (I) Data, (II) Model, (III) Training \& Inference. We systematically analyze interpretability from token-level to embedding-level representations, assess approaches related to both architecture analysis and design, and explore training and inference strategies that enhance transparency. By comparing various methodologies, we identify their strengths and limitations and propose future research directions to address unresolved challenges in multimodal explainability. This survey offers a foundational resource for advancing interpretability and transparency in MLLMs, guiding researchers and practitioners toward developing more accountable and robust multimodal AI systems. |
| title | Explainable and Interpretable Multimodal Large Language Models: A Comprehensive Survey |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2412.02104 |