Knowledge Acquisition Disentanglement for Knowledge-based Visual Question Answering with Large Language Models
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_ | 1866910536978923520 |
|---|---|
| author | An, Wenbin Tian, Feng Nie, Jiahao Shi, Wenkai Lin, Haonan Chen, Yan Wang, QianYing Wu, Yaqiang Dai, Guang Chen, Ping |
| author_facet | An, Wenbin Tian, Feng Nie, Jiahao Shi, Wenkai Lin, Haonan Chen, Yan Wang, QianYing Wu, Yaqiang Dai, Guang Chen, Ping |
| contents | Knowledge-based Visual Question Answering (KVQA) requires both image and world knowledge to answer questions. Current methods first retrieve knowledge from the image and external knowledge base with the original complex question, then generate answers with Large Language Models (LLMs). However, since the original question contains complex elements that require knowledge from different sources, acquiring different kinds of knowledge in a coupled manner may confuse models and hinder them from retrieving precise knowledge. Furthermore, the ``forward-only'' answering process fails to explicitly capture the knowledge needs of LLMs, which can further hurt answering quality. To cope with the above limitations, we propose DKA: Disentangled Knowledge Acquisition from LLM feedback, a training-free framework that disentangles knowledge acquisition to avoid confusion and uses LLM's feedback to specify the required knowledge. Specifically, DKA requires LLMs to specify what knowledge they need to answer the question and decompose the original complex question into two simple sub-questions: Image-based sub-question and Knowledge-based sub-question. Then we use the two sub-questions to retrieve knowledge from the image and knowledge base, respectively. In this way, two knowledge acquisition models can focus on the content that corresponds to them and avoid disturbance of irrelevant elements in the original complex question, which can help to provide more precise knowledge and better align the knowledge needs of LLMs to yield correct answers. Experiments on benchmark datasets show that DKA significantly outperforms SOTA models. To facilitate future research, our data and code are available at \url{https://github.com/Lackel/DKA}. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2407_15346 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Knowledge Acquisition Disentanglement for Knowledge-based Visual Question Answering with Large Language Models An, Wenbin Tian, Feng Nie, Jiahao Shi, Wenkai Lin, Haonan Chen, Yan Wang, QianYing Wu, Yaqiang Dai, Guang Chen, Ping Computer Vision and Pattern Recognition Computation and Language Multimedia Knowledge-based Visual Question Answering (KVQA) requires both image and world knowledge to answer questions. Current methods first retrieve knowledge from the image and external knowledge base with the original complex question, then generate answers with Large Language Models (LLMs). However, since the original question contains complex elements that require knowledge from different sources, acquiring different kinds of knowledge in a coupled manner may confuse models and hinder them from retrieving precise knowledge. Furthermore, the ``forward-only'' answering process fails to explicitly capture the knowledge needs of LLMs, which can further hurt answering quality. To cope with the above limitations, we propose DKA: Disentangled Knowledge Acquisition from LLM feedback, a training-free framework that disentangles knowledge acquisition to avoid confusion and uses LLM's feedback to specify the required knowledge. Specifically, DKA requires LLMs to specify what knowledge they need to answer the question and decompose the original complex question into two simple sub-questions: Image-based sub-question and Knowledge-based sub-question. Then we use the two sub-questions to retrieve knowledge from the image and knowledge base, respectively. In this way, two knowledge acquisition models can focus on the content that corresponds to them and avoid disturbance of irrelevant elements in the original complex question, which can help to provide more precise knowledge and better align the knowledge needs of LLMs to yield correct answers. Experiments on benchmark datasets show that DKA significantly outperforms SOTA models. To facilitate future research, our data and code are available at \url{https://github.com/Lackel/DKA}. |
| title | Knowledge Acquisition Disentanglement for Knowledge-based Visual Question Answering with Large Language Models |
| topic | Computer Vision and Pattern Recognition Computation and Language Multimedia |
| url | https://arxiv.org/abs/2407.15346 |