Co/plot: Supporting the Research Process Through Visualization

In the research process, data visualization plays a key role in understanding experimental results and deciding next steps. However, there was a bottleneck in that existing Matplotlib is difficult to iterate on repeatedly, while design tools like Figma struggle to guarantee data accuracy.

To solve this, co/plot was developed with the following characteristics.

• Fast prototyping: Provides predefined styles while still allowing users to freely customize them.
• Maintains data accuracy: Unlike design tools, it visualizes data while preserving the fidelity of the original data.
• Optimized research workflow: Supports iterative revision processes, helping researchers more clearly grasp the meaning of their data.

co/plot had its performance validated while processing data in over 70 languages during the development of the Tiny Aya model. In keeping with the spirit of open science, Cohere Labs has released this tool and plans to support researchers in sharing their findings faster and more accurately.