Pyramid Coder: Hierarchical Code Generator for Compositional Visual Question Answering

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Shen, Ruoyue, Inoue, Nakamasa, Shinoda, Koichi
Formato: Preprint
Publicado: 2024
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866910546131943424
author Shen, Ruoyue
Inoue, Nakamasa
Shinoda, Koichi
author_facet Shen, Ruoyue
Inoue, Nakamasa
Shinoda, Koichi
contents Visual question answering (VQA) is the task of providing accurate answers to natural language questions based on visual input. Programmatic VQA (PVQA) models have been gaining attention recently. These use large language models (LLMs) to formulate executable programs that address questions requiring complex visual reasoning. However, there are challenges in enabling LLMs to comprehend the usage of image processing modules and generate relevant code. To overcome these challenges, this paper introduces PyramidCoder, a novel prompting framework for PVQA models. PyramidCoder consists of three hierarchical levels, each serving a distinct purpose: query rephrasing, code generation, and answer aggregation. Notably, PyramidCoder utilizes a single frozen LLM and pre-defined prompts at each level, eliminating the need for additional training and ensuring flexibility across various LLM architectures. Compared to the state-of-the-art PVQA model, our approach improves accuracy by at least 0.5% on the GQA dataset, 1.4% on the VQAv2 dataset, and 2.9% on the NLVR2 dataset.
format Preprint
id arxiv_https___arxiv_org_abs_2407_20563
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Pyramid Coder: Hierarchical Code Generator for Compositional Visual Question Answering
Shen, Ruoyue
Inoue, Nakamasa
Shinoda, Koichi
Computer Vision and Pattern Recognition
Artificial Intelligence
Visual question answering (VQA) is the task of providing accurate answers to natural language questions based on visual input. Programmatic VQA (PVQA) models have been gaining attention recently. These use large language models (LLMs) to formulate executable programs that address questions requiring complex visual reasoning. However, there are challenges in enabling LLMs to comprehend the usage of image processing modules and generate relevant code. To overcome these challenges, this paper introduces PyramidCoder, a novel prompting framework for PVQA models. PyramidCoder consists of three hierarchical levels, each serving a distinct purpose: query rephrasing, code generation, and answer aggregation. Notably, PyramidCoder utilizes a single frozen LLM and pre-defined prompts at each level, eliminating the need for additional training and ensuring flexibility across various LLM architectures. Compared to the state-of-the-art PVQA model, our approach improves accuracy by at least 0.5% on the GQA dataset, 1.4% on the VQAv2 dataset, and 2.9% on the NLVR2 dataset.
title Pyramid Coder: Hierarchical Code Generator for Compositional Visual Question Answering
topic Computer Vision and Pattern Recognition
Artificial Intelligence
url https://arxiv.org/abs/2407.20563