Syon-Li/LongbenchSeg
Dataset Card for LongbenchSeg The segmented version of Longbench using the trained segmenter from the paper Towards Generalization of Block Attention via Automatic Segmentation and Block Distillation. We set the recursion depth to 1 and use the threshold value of 0.4. Dataset Details Dataset Description The newly introduced segmentation columns are: chunks: The segmented chunks. cut_prob: The corresponding segmenting probability for each… See the full description on the dataset page: https://huggingface.co/datasets/Syon-Li/LongbenchSeg.
Dataset Card for LongbenchSeg
<!-- Provide a quick summary of the dataset. -->
The segmented version of Longbench using the trained segmenter from the paper Towards Generalization of Block Attention via Automatic Segmentation and Block Distillation.
We set the recursion depth to 1 and use the threshold value of 0.4.
Dataset Details
Dataset Description
<!-- Provide a longer summary of what this dataset is. -->
The newly introduced segmentation columns are:
- chunks: The segmented chunks.
- cut_prob: The corresponding segmenting probability for each candidate cut point.
- parallel_degree: The number of segmented chunks.
If you find this dataset useful, please cite:
@article{DBLP:journals/corr/abs-2605-15913,
author = {Shuaiyi Li and
Zhisong Zhang and
Yan Wang and
Lei Zhu and
Dongyang Ma and
Chenlong Deng and
Yang Deng and
Wai Lam},
title = {Towards Generalization of Block Attention via Automatic Segmentation
and Block Distillation},
journal = {CoRR},
volume = {abs/2605.15913},
year = {2026},
url = {https://doi.org/10.48550/arXiv.2605.15913},
doi = {10.48550/ARXIV.2605.15913},
eprinttype = {arXiv},
eprint = {2605.15913},
timestamp = {Thu, 11 Jun 2026 10:32:45 +0200},
biburl = {https://dblp.org/rec/journals/corr/abs-2605-15913.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}