Back to Search Start Over

OmniBench: Towards The Future of Universal Omni-Language Models

Authors :
Li, Yizhi
Zhang, Ge
Ma, Yinghao
Yuan, Ruibin
Zhu, Kang
Guo, Hangyu
Liang, Yiming
Liu, Jiaheng
Yang, Jian
Wu, Siwei
Qu, Xingwei
Shi, Jinjie
Zhang, Xinyue
Yang, Zhenzhu
Wang, Xiangzhou
Zhang, Zhaoxiang
Liu, Zachary
Benetos, Emmanouil
Huang, Wenhao
Lin, Chenghua
Publication Year :
2024

Abstract

Recent advancements in multimodal large language models (MLLMs) have aimed to integrate and interpret data across diverse modalities. However, the capacity of these models to concurrently process and reason about multiple modalities remains inadequately explored, partly due to the lack of comprehensive modality-wise benchmarks. We introduce OmniBench, a novel benchmark designed to rigorously evaluate models' ability to recognize, interpret, and reason across visual, acoustic, and textual inputs simultaneously. We define models capable of such tri-modal processing as omni-language models (OLMs). OmniBench is distinguished by high-quality human annotations, ensuring that accurate responses require integrated understanding and reasoning across all three modalities. Our main findings reveal that: i) most OLMs exhibit critical limitations in instruction-following and reasoning capabilities within tri-modal contexts; and ii) most baselines models perform poorly (below 50\% accuracy) even when provided with alternative textual representations of images or/and audio. These results suggest that the ability to construct a consistent context from text, image, and audio is often overlooked in existing MLLM training paradigms. We advocate for future research to focus on developing more robust tri-modal integration techniques and training strategies to enhance OLM performance across diverse modalities. The codes and live leaderboard could be found at https://m-a-p.ai/OmniBench.

Details

Database :
arXiv
Publication Type :
Report
Accession number :
edsarx.2409.15272
Document Type :
Working Paper