中文 ｜ English
**TOUCHSTONE** is a comprehensive assessment of multimodal language models, encompassing not only basic recognition and comprehension but also extending to literary creation. By automating the evaluation process and converting multimodal information into text, our TouchStone allows for efficient and accurate assessment of dialogue quality, leveraging the power of advanced language models without the need for manual intervention.
## DATASET
To evaluate the abilities of LVLMs, we construct a diverse and comprehensive dataset that covers five key dimensions: basic descriptive ability, visual recognition ability, visual comprehension ability, visual storytelling ability, and multi-image analysis ability.
- **Basic Descriptive Ability** Image description involves the ability of a model to describe the information contained in an image, including simple and detailed descriptions. Simple descriptions are typically short phrases that describe the main subject and action of the image, while detailed descriptions provide more in-depth information about the image scene, their attributes, and relationships.
- **Visual Recognition Ability** Image recognition is the task of recognizing objects or scenes within an image and inferring relevant information. This area can be further divided into several sub-tasks, including attribute QA, movie/TV recognition, art recognition, landmark recognition, celebrity recognition, emotion recognition, text recognition, object recognition, and structure content recognition.
- **Visual Comprehension Ability** Image understanding involves the ability of a model to understand the meaning of an image and associated tasks. This area encompasses several sub-tasks, such as style appreciation, abstract image understanding, meme understanding, image analysis, chart analysis, general problem-solving, and reasoning QA.
- **Visual Storytelling Ability** The visual storytelling ability is the process of literary creation based on visual content, including writing emails, poetry, stories, ads/commodity recommendations, and brainstorming.
- **Multi-Image Analysis Ability** Multi-image analysis is the task of analyzing and comparing multiple images. This area includes tasks such as comparing two/multiple images, summarizing multiple image information, comparing commodities, and step-by-step analysis of images.
We comprehensively evaluate the model's ability from five dimensions. As shown in the figure above, an example of 27 subtasks is given. From perception to cognition to creativity, as the difficulty increases, the requirements for models are also getting higher and higher. Currently, LVLM capabilities are in their early stages. Our dataset contains 800+ questions and 27 categories.
## Methods
We apply a powerful LLM as a judge to enable automated evaluation. To effectively comprehend the contents of an image, we manually substitute the actual image input with fine-grained textual annotations. By inputting these annotations and corresponding questions to a powerful LLM like GPT4, we obtain reference answers.
For the evaluation of the LVLMs, we provide actual images and questions as input and obtain their respective answers. Finally, we employ GPT4 to score the answers generated by the LVLMs based on the fine-grained annotations and questions. The scoring instructions require the model to assess the usefulness, relevance, and accuracy of the answers, considering the annotations as the content of the images. To ensure fairness in the evaluation, each model's answer is compared against a consistent reference answer from GPT4. The average score of the model in all questions is taken as the final score.
To eliminate the influence of answer position, we perform a second scoring round by swapping the positions of the answers and then compute the average of the two scores obtained. This approach aims to mitigate any bias introduced by the placement of the answers.
### Evaluation
#### Evaluation in English-based Multimodal Dialogue
| Model | Score |
|---------------|-------|
| PandaGPT | 488.5 |
| MiniGPT4 | 531.7 |
| InstructBLIP | 552.4 |
| LLaMA-AdapterV2 | 590.1 |
| mPLUG-Owl | 605.4 |
| LLaVA | 602.7 |
| Qwen-VL-Chat | 645.2 |
#### Evaluation in Chinese-based Multimodal Dialogue
| Model | Score |
|---------------|-------|
| VisualGLM | 247.1 |
| Qwen-VL-Chat | 401.2 |