$FastDoc$: Domain-Specific Fast Continual Pre-training Technique using Document-Level Metadata and Taxonomy

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Nandy, Abhilash, Kapadnis, Manav Nitin, Patnaik, Sohan, Butala, Yash Parag, Goyal, Pawan, Ganguly, Niloy
Natura: Preprint
Pubblicazione: 2023
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866912098691317760
author Nandy, Abhilash
Kapadnis, Manav Nitin
Patnaik, Sohan
Butala, Yash Parag
Goyal, Pawan
Ganguly, Niloy
author_facet Nandy, Abhilash
Kapadnis, Manav Nitin
Patnaik, Sohan
Butala, Yash Parag
Goyal, Pawan
Ganguly, Niloy
contents In this paper, we propose $FastDoc$ (Fast Continual Pre-training Technique using Document Level Metadata and Taxonomy), a novel, compute-efficient framework that utilizes Document metadata and Domain-Specific Taxonomy as supervision signals to continually pre-train transformer encoder on a domain-specific corpus. The main innovation is that during domain-specific pretraining, an open-domain encoder is continually pre-trained using sentence-level embeddings as inputs (to accommodate long documents), however, fine-tuning is done with token-level embeddings as inputs to this encoder. We perform such domain-specific pre-training on three different domains namely customer support, scientific, and legal domains, and compare performance on 6 different downstream tasks and 9 different datasets. The novel use of document-level supervision along with sentence-level embedding input for pre-training reduces pre-training compute by around $1,000$, $4,500$, and $500$ times compared to MLM and/or NSP in Customer Support, Scientific, and Legal Domains, respectively. The reduced training time does not lead to a deterioration in performance. In fact we show that $FastDoc$ either outperforms or performs on par with several competitive transformer-based baselines in terms of character-level F1 scores and other automated metrics in the Customer Support, Scientific, and Legal Domains. Moreover, reduced training aids in mitigating the risk of catastrophic forgetting. Thus, unlike baselines, $FastDoc$ shows a negligible drop in performance on open domain.
format Preprint
id arxiv_https___arxiv_org_abs_2306_06190
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle $FastDoc$: Domain-Specific Fast Continual Pre-training Technique using Document-Level Metadata and Taxonomy
Nandy, Abhilash
Kapadnis, Manav Nitin
Patnaik, Sohan
Butala, Yash Parag
Goyal, Pawan
Ganguly, Niloy
Computation and Language
Machine Learning
68T50
I.2.7
In this paper, we propose $FastDoc$ (Fast Continual Pre-training Technique using Document Level Metadata and Taxonomy), a novel, compute-efficient framework that utilizes Document metadata and Domain-Specific Taxonomy as supervision signals to continually pre-train transformer encoder on a domain-specific corpus. The main innovation is that during domain-specific pretraining, an open-domain encoder is continually pre-trained using sentence-level embeddings as inputs (to accommodate long documents), however, fine-tuning is done with token-level embeddings as inputs to this encoder. We perform such domain-specific pre-training on three different domains namely customer support, scientific, and legal domains, and compare performance on 6 different downstream tasks and 9 different datasets. The novel use of document-level supervision along with sentence-level embedding input for pre-training reduces pre-training compute by around $1,000$, $4,500$, and $500$ times compared to MLM and/or NSP in Customer Support, Scientific, and Legal Domains, respectively. The reduced training time does not lead to a deterioration in performance. In fact we show that $FastDoc$ either outperforms or performs on par with several competitive transformer-based baselines in terms of character-level F1 scores and other automated metrics in the Customer Support, Scientific, and Legal Domains. Moreover, reduced training aids in mitigating the risk of catastrophic forgetting. Thus, unlike baselines, $FastDoc$ shows a negligible drop in performance on open domain.
title $FastDoc$: Domain-Specific Fast Continual Pre-training Technique using Document-Level Metadata and Taxonomy
topic Computation and Language
Machine Learning
68T50
I.2.7
url https://arxiv.org/abs/2306.06190