The prompt for extracting figures supports Chinese.

[liuzhenghua]

目前提取表格的步骤，使用的是英文prompt，没有根据文档的语言去支持中文，导致提取的结果中英混合，理论上有可能影响模型回答的效果。
Currently, the table extraction process uses English prompts without adapting to the language of the document. As a result, the extracted content is a mix of Chinese and English, which could theoretically affect the model's response quality.

Here are some of my test cases：

prompt:

You are an expert visual data analyst. Analyze the image and provide a comprehensive description of its content. Focus on identifying the type of visual data representation (e.g., bar chart, pie chart, line graph, table, flowchart), its structure, and any text captions or labels included in the image.

Tasks:
1. Describe the overall structure of the visual representation. Specify if it is a chart, graph, table, or diagram.
2. Identify and extract any axes, legends, titles, or labels present in the image. Provide the exact text where available.
3. Extract the data points from the visual elements (e.g., bar heights, line graph coordinates, pie chart segments, table rows and columns).
4. Analyze and explain any trends, comparisons, or patterns shown in the data.
5. Capture any annotations, captions, or footnotes, and explain their relevance to the image.
6. Only include details that are explicitly present in the image. If an element (e.g., axis, legend, or caption) does not exist or is not visible, do not mention it.

Output format (include only sections relevant to the image content):
- Visual Type: [Type]
- Title: [Title text, if available]
- Axes / Legends / Labels: [Details, if available]
- Data Points: [Extracted data]
- Trends / Insights: [Analysis and interpretation]
- Captions / Annotations: [Text and relevance, if available]

Ensure high accuracy, clarity, and completeness in your analysis, and includes only the information present in the image. Avoid unnecessary statements about missing elements.

case 1

Image

result

### Visual Type: Bar Chart with Line Graph Overlay

### Title: 
1. 定义及现状 (Definition and Current Status)
### Subtitle:
二甲基亚砜 (DMSO) 行业定义及现状 (DMSO Industry Definition and Current Status)

### Axes / Legends / Labels:
- **X-axis:** Years from 2017 to 2023.
- **Y-axis (Left):** Quantity in "万噸" (ten thousand tons), ranging from 0 to 8.
- **Y-axis (Right):** Market scale in "亿元" (billion yuan), ranging from 0 to 20.
- **Legend:**
  - 产量 (Production) - Dark blue bars.
  - 需求量 (Demand) - Light blue bars.
  - 市场规模 (Market Scale) - Light blue line graph.

### Data Points:
- **2017:** Production = 5.7, Demand = 3.3, Market Scale = 4.44.
- **2018:** Production = 5.8, Demand = 3.3, Market Scale = 3.3.
- **2019:** Production = 5.5, Demand = 3.0, Market Scale = 2.8.
- **2020:** Production = 5.5, Demand = 3.0, Market Scale = 2.7.
- **2021:** Production = 5.7, Demand = 4.0, Market Scale = 6.0.
- **2022:** Production = 6.0, Demand = 4.4, Market Scale = 14.0.
- **2023:** Production = 6.0, Demand = 4.85, Market Scale = 18.19.

### Trends / Insights:
- The production of DMSO remained relatively stable from 2017 to 2023, fluctuating between 5.5 and 6.0万噸.
- The demand for DMSO showed a gradual increase, rising from 3.3万噸 in 2017 to 4.85万噸 in 2023.
- The market scale experienced significant growth, increasing from 4.44 billion yuan in 2017 to 18.19 billion yuan in 2023, indicating a strong expansion in the industry.

### Captions / Annotations:
- **Top Right Annotation:** Highlights the increase in demand from 3.3万噸 in 2017 to 4.85万噸 in 2023.
- **Bottom Right Annotation:** Shows the market scale growth from 4.44 billion yuan in 2017 to 18.19 billion yuan in 2023.
- **Footnote:** "资料来源：公开资料、智研咨询整理" (Data Source: Public Data, Organized by Zhiyan Consulting).二甲基亚砜（DMSO）具有高极性、高沸点、热稳定性好、非质子、与水混溶的特性，既是一种氢键破坏剂，还是一种具有吸湿性的可燃液体，因其能溶于乙醇、丙醇、苯和氯仿等大多数有机物，被誉为“万能”溶剂，广泛应用于农药、医药、集成电路、碳纤维等领域，随着下游市场的飞速发展，带动二甲基亚砜市场需求快速增长，市场规模不断扩大，2023年我国二甲基亚砜需求量达4.85万吨，市场规模达18.19亿元。

case 2

image

result

## Visual Type
The visual representation is an infographic that includes a map and accompanying text.

## Title
The title of the infographic is "我国二甲基亚砜（DMSO）行业主要厂商分布地图," which translates to "China's Dimethyl Sulfoxide (DMSO) Industry Major Manufacturers Distribution Map."

## Axes / Legends / Labels
- The map is labeled with the names of companies and their corresponding locations.
- The companies listed are:
  1. 新疆兴发 (Xinjiang Xingfa)
  2. 阳煤丰喜 (Yangmei Fengxi)
  3. 沧州东丽 (Cangzhou Dongli)
  4. 湖北兴发集团 (Hubei Xingfa Group)
  5. 重庆兴发 (Chongqing Xingfa)
  6. 湖南汉森化工 (Hunan Hansen Chemical)

## Data Points
The infographic highlights six major manufacturers of DMSO in China, located in different regions:
- 新疆兴发 (Xinjiang Xingfa) in Xinjiang
- 阳煤丰喜 (Yangmei Fengxi) in Shanxi
- 沧州东丽 (Cangzhou Dongli) in Hebei
- 湖北兴发集团 (Hubei Xingfa Group) in Hubei
- 重庆兴发 (Chongqing Xingfa) in Chongqing
- 湖南汉森化工 (Hunan Hansen Chemical) in Hunan

## Trends / Insights
The infographic indicates that the DMSO industry in China is distributed across various regions, with a concentration of major manufacturers in different provinces. The text accompanying the map provides additional information about the industry and specific companies.

## Captions / Annotations
The infographic includes three main sections of text that provide detailed information about the following companies:
1. 湖北兴发化工集团股份有限公司 (Hubei Xingfa Chemical Group Co., Ltd.)
2. 重庆兴发金冠化工有限公司 (Chongqing Xingfa JinGuan Chemical Co., Ltd.)
3. 沧州东丽精细化工有限公司 (Cangzhou Dongli Fine Chemical Co., Ltd.)

These sections highlight the business scope, financial performance, and other relevant details of the mentioned companies. The footer of the infographic notes that the data sources include public materials and information organized by 智研咨询 (Qianzhan Research).目前，国内二甲基亚砜行业生产企业数量不多，大大小小的企业才十来家，行业产能较大的企业主要包括重庆兴发、新疆兴发、沧州东丽、阳煤丰喜、汉森化工等，其中，兴发集团作为二甲基亚砜龙头，多年来始终坚持“精细化、绿色化、高端化、国际化”发展思路，持续加强自主创新能力建设，先后组织实施了40多个国家、省市级重点科技计划项目；“二甲基亚砜绿色高效合成关键技术及产业化”荣获中国石油和化学工业科技进步一等奖；参与制定国际标准1项，主持和参与制定国家、行业及团体各类标准84项。

[ZhenhangTung]

@liuzhenghua Thanks so much for the feedback — really appreciate it!
I believe our prompt will remain in English by default to ensure consistency and broader model compatibility.
That said, we’re definitely open to discussing how to best support your goal @KevinHuSh @yongtenglei

[ZhenhangTung]

@liuzhenghua Thanks so much for the feedback — really appreciate it!
I believe our prompt will remain in English by default to ensure consistency and broader model compatibility.
That said, we’re definitely open to discussing how to best support your goal @KevinHuSh @yongtenglei