Skip to content

MyGO

Complete multimodal knowledge graphs with fine-grained tokens

Visit official site

What MyGO is

The tool is described as a multimodal knowledge graph completion method that handles discrete modality information as fine-grained tokens to raise completion accuracy. It embeds text tokens with the transformers library before training and evaluation on multimodal datasets.

What you can do with it

  • Embed text tokens
  • Train on multimodal datasets
  • Support custom datasets
  • Reproduce experiment results

How it works

  • Discrete modality information is treated as fine-grained tokens
  • Text tokens are embedded with the transformers library
  • Training scripts are provided to reproduce experiment results

Pros & cons

✓ What we like

  • Custom datasets are supported
  • Training and evaluation cover multimodal datasets

! What to watch out for

  • Text token embedding relies on the transformers library
  • The described workflow centres on multimodal datasets

FAQ

What does the tool complete?

The material describes multimodal knowledge graph completion.

Can I use my own data?

Custom dataset support and training scripts for reproducing results are listed.

Last reviewed: 2026-11-15

More AI office assistant tools

View all →

How we review