OmniBench: Towards The Future of Universal Omni-Language Models
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , , , , , , , , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2024
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866918265482117120 |
|---|---|
| author | Li, Yizhi Ma, Yinghao Zhang, Ge Yuan, Ruibin Zhu, Kang Guo, Hangyu Liang, Yiming Liu, Jiaheng Wang, Zekun Yang, Jian Wu, Siwei Qu, Xingwei Shi, Jinjie Zhang, Xinyue Yang, Zhenzhu Wen, Yidan Wang, Yanghai Li, Shihao Zhang, Zhaoxiang Liu, Zachary Benetos, Emmanouil Huang, Wenhao Lin, Chenghua |
| author_facet | Li, Yizhi Ma, Yinghao Zhang, Ge Yuan, Ruibin Zhu, Kang Guo, Hangyu Liang, Yiming Liu, Jiaheng Wang, Zekun Yang, Jian Wu, Siwei Qu, Xingwei Shi, Jinjie Zhang, Xinyue Yang, Zhenzhu Wen, Yidan Wang, Yanghai Li, Shihao Zhang, Zhaoxiang Liu, Zachary Benetos, Emmanouil Huang, Wenhao Lin, Chenghua |
| contents | 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 underexplored, 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 language models capable of such tri-modal processing as the 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) open-source 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. To address this gap, we curate an instruction tuning dataset of 84.5K training samples, OmniInstruct, for training OLMs to adapt to tri-modal contexts. We advocate for future research to focus on developing more robust tri-modal integration techniques and training strategies to enhance OLMs. Codes, data and live leaderboard could be found at https://m-a-p.ai/OmniBench. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2409_15272 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | OmniBench: Towards The Future of Universal Omni-Language Models Li, Yizhi Ma, Yinghao Zhang, Ge Yuan, Ruibin Zhu, Kang Guo, Hangyu Liang, Yiming Liu, Jiaheng Wang, Zekun Yang, Jian Wu, Siwei Qu, Xingwei Shi, Jinjie Zhang, Xinyue Yang, Zhenzhu Wen, Yidan Wang, Yanghai Li, Shihao Zhang, Zhaoxiang Liu, Zachary Benetos, Emmanouil Huang, Wenhao Lin, Chenghua Computation and Language Artificial Intelligence Computer Vision and Pattern Recognition 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 underexplored, 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 language models capable of such tri-modal processing as the 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) open-source 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. To address this gap, we curate an instruction tuning dataset of 84.5K training samples, OmniInstruct, for training OLMs to adapt to tri-modal contexts. We advocate for future research to focus on developing more robust tri-modal integration techniques and training strategies to enhance OLMs. Codes, data and live leaderboard could be found at https://m-a-p.ai/OmniBench. |
| title | OmniBench: Towards The Future of Universal Omni-Language Models |
| topic | Computation and Language Artificial Intelligence Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2409.15272 |