Paper ID: 2211.17135

BudgetLongformer: Can we Cheaply Pretrain a SotA Legal Language Model From Scratch?

Joel Niklaus, Daniele Giofré

Pretrained transformer models have achieved state-of-the-art results in many tasks and benchmarks recently. Many state-of-the-art Language Models (LMs), however, do not scale well above the threshold of 512 input tokens. In specialized domains though (such as legal, scientific or biomedical), models often need to process very long text (sometimes well above 10000 tokens). Even though many efficient transformers have been proposed (such as Longformer, BigBird or FNet), so far, only very few such efficient models are available for specialized domains. Additionally, since the pretraining process is extremely costly in general - but even more so as the sequence length increases - it is often only in reach of large research labs. One way of making pretraining cheaper is the Replaced Token Detection (RTD) task, by providing more signal during training, since the loss can be computed over all tokens. In this work, we train Longformer models with the efficient RTD task on legal data to showcase that pretraining efficient LMs is possible using much less compute. We evaluate the trained models on challenging summarization tasks requiring the model to summarize long texts to show to what extent the models can achieve good performance on downstream tasks. We find that both the small and base models outperform their baselines on the in-domain BillSum and out-of-domain PubMed tasks in their respective parameter range. We publish our code and models for research purposes.

Submitted: Nov 30, 2022