google-research-datasets/xsum_factuality
Neural abstractive summarization models are highly prone to hallucinate content that is unfaithful to the input document. The popular metric such as ROUGE fails to show the severity of the problem. The dataset consists of faithfulness and factuality annotations of abstractive summaries for the XSum dataset. We have crowdsourced 3 judgements for each of 500 x 5 document-system pairs. This will be a valuable resource to the abstractive summarization community.
| Author / organization | google-research-datasets |
|---|---|
| License (as declared) | cc-by-4.0 |
| Downloads | 259 (as reported by Hugging Face) |
| Likes | 6 |
| Created | 2 March 2022 |
| Last updated at source | 18 January 2024 |
Popularity, last 3 days
259 now (-1 since 2026-10-16). Counts as reported by the source, snapshotted daily by Silytics.
More from google-research-datasets
Silytics shows public metadata from Hugging Face and links to the original. Information may be incomplete or out of date. Last refreshed 18 October 2026. Licenses and terms of the original apply to the work itself. Not financial, legal or professional advice.