Automatic Evaluation Results for CNN/DailyMail Summarization Models

Evaluation results from an automated benchmarking project for text summarization models on the CNN/DailyMail dataset. The project was created by autoevaluate and last updated in July 2022. The specific scope and size of the results are not detailed.

Use Cases

• Compare model performance on standard ROUGE or BLEU metrics using the provided evaluation scores.
• Benchmark new summarization models against results from the autoevaluate staging project.
• Analyze potential correlations between different automated evaluation metrics reported in the results.

Strengths

• Results are tied to the standard CNN/DailyMail benchmark for summarization.
• Project was updated in 2022, indicating a defined point of comparison.

Limitations

• The exact number of evaluated model runs or result rows is unknown.
• The specific evaluation metrics and their values are not provided in the description.
• The data may be stale for comparing against models developed after 2022.

Provenance

Source

autoevaluate

Collection Method

Automated evaluation of models on the CNN/DailyMail dataset.

Freshness

Last updated 2022-07-27.