https://github.com/webis-de/summary-workbench
Framework for unified summarisation and evaluation of English documents using state-of-the-art models and measures.
https://github.com/webis-de/summary-workbench
evaluation-metrics summarization text-generation visualization
Last synced: 6 months ago
JSON representation
Framework for unified summarisation and evaluation of English documents using state-of-the-art models and measures.
- Host: GitHub
- URL: https://github.com/webis-de/summary-workbench
- Owner: webis-de
- License: mit
- Created: 2022-07-22T11:54:49.000Z (about 3 years ago)
- Default Branch: main
- Last Pushed: 2024-05-13T18:56:15.000Z (over 1 year ago)
- Last Synced: 2025-03-22T20:26:15.006Z (7 months ago)
- Topics: evaluation-metrics, summarization, text-generation, visualization
- Language: Python
- Homepage: https://tldr.demo.webis.de
- Size: 81.2 MB
- Stars: 32
- Watchers: 13
- Forks: 7
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# Summary Workbench
Unifying the application and evaluation of text summarization models. [[Paper](https://arxiv.org/pdf/2210.09587.pdf)] [[Documentation](https://webis-de.github.io/summary-workbench/)] [[Live Demo](https://tldr.demo.webis.de)]
>Accepted at EMNLP 2022 (Demo track). :tada: :tada:#### :loudspeaker: Updates (3-12-2022)
1. Integrated 2 new models and their variants (6 in total): [BRIO](https://arxiv.org/abs/2203.16804) trained on news, [Schnitsum](https://github.com/sobamchan/schnitsum) trained on scholary documents.
2. Integrated [contrastive search](https://huggingface.co/blog/introducing-csearch) for more fluent summaries. User can now toggle between regular and contrasitve search for supported models.
2. Improvements to the UI responsiveness on smaller devices.## Summarize
### Create a Request

### Inspect the Results

# Evaluate
### Create a Request

### Inspect the Results
#### Scores

#### Visualize Text Examples

#### Plot Scores against each other
