AutoTrain Summarization Predictions on CNN DailyMail Test Set

Featuring model predictions for a summarization task on the CNN/DailyMail dataset's test split. The predictions were generated by the pszemraj/long-t5-tglobal-base-16384-book-summary model via the AutoTrain platform. The row count, column count, and specific data fields are unknown.

Use Cases

• Compare generated summaries against reference texts to evaluate summarization model performance.
• Analyze the output of the long-t5-tglobal-base-16384-book-summary model on a standard news summarization benchmark.
• Use the predictions as a baseline for developing new automatic summarization evaluation metrics.

Strengths

• Predictions are for a standard, widely-used benchmark dataset (CNN/DailyMail).
• Generated by a specific, named model architecture (long-t5-tglobal-base-16384-book-summary).
• Targets a defined task (Summarization) and dataset split (test).

Limitations

• The specific data columns, sample data, and total row count are unknown.
• The dataset's last known update was in July 2022, which may be stale for current model comparisons.
• Without the original CNN/DailyMail articles or reference summaries, the predictions cannot be fully evaluated in isolation.

Provenance

Source

Hugging Face AutoTrain Evaluator

Collection Method

Model inference outputs generated by the AutoTrain platform.

Freshness

2022-07-16