LongBlocks
- Type
- dataset
- Venue
- utter-project
- Year
- 2026
- Source
- huggingface
- Access
- free
- Language
- en, fr, de, es, uk, sv, ro, it, ru, el, ja, nl, fi, pl, hu, zh, pt, hi, ar
- Added
- 2026-07-17T20:02:33.512080+00:00
- Verified
- 2026-07-17T20:02:33.512080+00:00
Summary
The dataset was created to support long-context adaptation for tasks that require reasoning over extended inputs, including:
Keywords
hf-dataset language-modeling · -qa parquet optimized-parquet text datasets dask polars mlcroissant has-paper
Topics
Language Modeling, QA
Research notes
- downloads=829; likes=8