Saved in:
Bibliographic Details
Main Authors: Huang, Yupan, Meng, Zaiqiao, Liu, Fangyu, Su, Yixuan, Collier, Nigel, Lu, Yutong
Format: Preprint
Published: 2023
Subjects:
Online Access:https://arxiv.org/abs/2308.16463
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910605971030016
author Huang, Yupan
Meng, Zaiqiao
Liu, Fangyu
Su, Yixuan
Collier, Nigel
Lu, Yutong
author_facet Huang, Yupan
Meng, Zaiqiao
Liu, Fangyu
Su, Yixuan
Collier, Nigel
Lu, Yutong
contents Large language models exhibit enhanced zero-shot performance on various tasks when fine-tuned with instruction-following data. Multimodal instruction-following models extend these capabilities by integrating both text and images. However, existing models such as MiniGPT-4 and LLaVA face challenges in maintaining dialogue coherence in scenarios involving multiple images. A primary reason is the lack of a specialized dataset for this critical application. To bridge these gaps, we introduce SparklesDialogue, the first machine-generated dialogue dataset tailored for word-level interleaved multi-image and text interactions. Furthermore, we construct SparklesEval, a GPT-assisted benchmark for quantitatively assessing a model's conversational competence across multiple images and dialogue turns. We then present SparklesChat, a multimodal instruction-following model for open-ended dialogues across multiple images. Our experiments validate the effectiveness of training SparklesChat with SparklesDialogue based on MiniGPT-4 and LLaVA-v1.5, which enhances comprehension across multiple images and dialogue turns, and does not compromise single-image understanding capabilities. Qualitative evaluations further demonstrate SparklesChat's generality in handling real-world applications. All resources related to this study are publicly available at https://github.com/HYPJUDY/Sparkles.
format Preprint
id arxiv_https___arxiv_org_abs_2308_16463
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Sparkles: Unlocking Chats Across Multiple Images for Multimodal Instruction-Following Models
Huang, Yupan
Meng, Zaiqiao
Liu, Fangyu
Su, Yixuan
Collier, Nigel
Lu, Yutong
Computer Vision and Pattern Recognition
Computation and Language
Large language models exhibit enhanced zero-shot performance on various tasks when fine-tuned with instruction-following data. Multimodal instruction-following models extend these capabilities by integrating both text and images. However, existing models such as MiniGPT-4 and LLaVA face challenges in maintaining dialogue coherence in scenarios involving multiple images. A primary reason is the lack of a specialized dataset for this critical application. To bridge these gaps, we introduce SparklesDialogue, the first machine-generated dialogue dataset tailored for word-level interleaved multi-image and text interactions. Furthermore, we construct SparklesEval, a GPT-assisted benchmark for quantitatively assessing a model's conversational competence across multiple images and dialogue turns. We then present SparklesChat, a multimodal instruction-following model for open-ended dialogues across multiple images. Our experiments validate the effectiveness of training SparklesChat with SparklesDialogue based on MiniGPT-4 and LLaVA-v1.5, which enhances comprehension across multiple images and dialogue turns, and does not compromise single-image understanding capabilities. Qualitative evaluations further demonstrate SparklesChat's generality in handling real-world applications. All resources related to this study are publicly available at https://github.com/HYPJUDY/Sparkles.
title Sparkles: Unlocking Chats Across Multiple Images for Multimodal Instruction-Following Models
topic Computer Vision and Pattern Recognition
Computation and Language
url https://arxiv.org/abs/2308.16463