# Simple Hack for Transformers against Heavy Long-Text Classification on a Time- and Memory-Limited GPU Service

arXiv: https://arxiv.org/abs/2403.12563 · HTML: https://mlsanity.com/paper/2403.12563/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 21 of its 36 references that are in the map.
- First posted on arXiv 19 Mar 2024.

## Builds on (most cited first, 21 of 21)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [DistilBERT, a distilled version of BERT: smaller, faster, cheaper and lighter](https://mlsanity.com/paper/1910.01108/) (2019)
- [Unsupervised Cross-lingual Representation Learning at Scale](https://mlsanity.com/paper/1911.02116/) (2019)
- [Longformer: The Long-Document Transformer](https://mlsanity.com/paper/2004.05150/) (2020)
- [ALBERT: A Lite BERT for Self-supervised Learning of Language Representations](https://mlsanity.com/paper/1909.11942/) (2019)
- [Optuna: A Next-generation Hyperparameter Optimization Framework](https://mlsanity.com/paper/1907.10902/) (2019)
- [Efficient Transformers: A Survey](https://mlsanity.com/paper/2009.06732/) (2020)
- [Nystr"omformer: A Nystr"om-Based Algorithm for Approximating Self-Attention](https://mlsanity.com/paper/2102.03902/) (2021)
- [How to Fine-Tune BERT for Text Classification?](https://mlsanity.com/paper/1905.05583/) (2019)
- [Hyperparameter Optimization: Foundations, Algorithms, Best Practices and Open Challenges](https://mlsanity.com/paper/2107.05847/) (2021)
- [Hyper-Parameter Optimization: A Review of Algorithms and Applications](https://mlsanity.com/paper/2003.05689/) (2020)
- [Hierarchical Transformers for Long Document Classification](https://mlsanity.com/paper/1910.10781/) (2019)
- [XLM-V: Overcoming the Vocabulary Bottleneck in Multilingual Masked Language Models](https://mlsanity.com/paper/2301.10472/) (2023)
- [NusaX: Multilingual Parallel Sentiment Dataset for 10 Indonesian Local Languages](https://mlsanity.com/paper/2205.15960/) (2022)
- [One Country, 700+ Languages: NLP Challenges for Underrepresented Languages and Dialects in Indonesia](https://mlsanity.com/paper/2203.13357/) (2022)
- [IndoLEM and IndoBERT: A Benchmark Dataset and Pre-trained Language Model for Indonesian NLP](https://mlsanity.com/paper/2011.00677/) (2020)
- [Efficient Classification of Long Documents Using Transformers](https://mlsanity.com/paper/2203.11258/) (2022)
- [IndoSum: A New Benchmark Dataset for Indonesian Text Summarization](https://mlsanity.com/paper/1810.05334/) (2018)
- [Which Student is Best? A Comprehensive Knowledge Distillation Exam for Task-Specific BERT Models](https://mlsanity.com/paper/2201.00558/) (2022)
- [Indonesia's Fake News Detection using Transformer Network](https://mlsanity.com/paper/2107.06796/) (2021)

## Abstract (arXiv)

Many NLP researchers rely on free computational services, such as Google Colab, to fine-tune their Transformer models, causing a limitation for hyperparameter optimization (HPO) in long-text classification due to the method having quadratic complexity and needing a bigger resource. In Indonesian, only a few works were found on long-text classification using Transformers. Most only use a small amount of data and do not report any HPO. In this study, using 18k news articles, we investigate which pretrained models are recommended to use based on the output length of the tokenizer. We then compare some hacks to shorten and enrich the sequences, which are the removals of stopwords, punctuation, low-frequency words, and recurring words. To get a fair comparison, we propose and run an efficient and dynamic HPO procedure that can be done gradually on a limited resource and does not require a long-running optimization library. Using the best hack found, we then compare 512, 256, and 128 tokens length. We find that removing stopwords while keeping punctuation and low-frequency words is the best hack. Some of our setups manage to outperform taking 512 first tokens using a smaller 128 or 256 first tokens which manage to represent the same information while requiring less computational resources. The findings could help developers to efficiently pursue optimal performance of the models using limited resources.