The model has been warm-started re-using the weights of RoBERTa (Liu et al., 2019), BUT has not been continued pre-trained. It supports sequences of length up to 4,096.

HAT uses hierarchical attention, which is a combination of segment-wise and cross-segment attention operations. You can think of segments as paragraphs or sentences.

Note: If you wish to use a fully pre-trained HAT model, you have to use kiddothe2b/adhoc-hat-base-4096.

Intended uses & limitations

The model is intended to be fine-tuned on a downstream task. See the model hub to look for other versions of HAT, or fine-tuned versions on a task that interests you.

Note that this model is primarily aimed at being fine-tuned on tasks that use the whole document to make decisions, such as document classification, sequential sentence classification, or question answering.

How to use

You can fine-tune it for SequenceClassification, SequentialSentenceClassification, and MultipleChoice down-stream tasks:

python

from transformers import AutoTokenizer, AutoModelForSequenceClassification
tokenizer = AutoTokenizer.from_pretrained("kiddothe2b/adhoc-hierarchical-transformer-base-4096", trust_remote_code=True)
doc_classifier = AutoModelForSequenceClassification("kiddothe2b/adhoc-hierarchical-transformer-base-4096", trust_remote_code=True)

Note: If you wish to use a fully pre-trained HAT model, you have to use kiddothe2b/hierarchical-transformer-base-4096.

Limitations and bias

The training data used for this model contains a lot of unfiltered content from the internet, which is far from neutral. Therefore, the model can have biased predictions.

Training procedure

Training and evaluation data

The model has been warm-started from roberta-base checkpoint.

Framework versions

Transformers 4.19.0.dev0
Pytorch 1.11.0+cu102
Datasets 2.0.0
Tokenizers 0.11.6

Citing

If you use HAT in your research, please cite:

An Exploration of Hierarchical Attention Transformers for Efficient Long Document Classification. Ilias Chalkidis, Xiang Dai, Manos Fergadiotis, Prodromos Malakasiotis, and Desmond Elliott. 2022. arXiv:2210.05529 (Preprint).

@misc{chalkidis-etal-2022-hat,
  url = {https://arxiv.org/abs/2210.05529},
  author = {Chalkidis, Ilias and Dai, Xiang and Fergadiotis, Manos and Malakasiotis, Prodromos and Elliott, Desmond},
  title = {An Exploration of Hierarchical Attention Transformers for Efficient Long Document Classification},
  publisher = {arXiv},
  year = {2022},
}