# Non-Contextual BERT or FastText? A Comparative Analysis

arXiv: https://arxiv.org/abs/2411.17661 · HTML: https://mlsanity.com/paper/2411.17661/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 9 of its 23 references that are in the map.
- First posted on arXiv 26 Nov 2024.

## Builds on (most cited first, 9 of 9)

- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [Efficient Estimation of Word Representations in Vector Space](https://mlsanity.com/paper/1301.3781/) (2013)
- [MuRIL: Multilingual Representations for Indian Languages](https://mlsanity.com/paper/2103.10730/) (2021)
- [L3Cube-MahaCorpus and MahaBERT: Marathi Monolingual Corpus, Marathi BERT Language Models, and Resources](https://mlsanity.com/paper/2202.01159/) (2022)
- [L3Cube-MahaHate: A Tweet-based Marathi Hate Speech Detection Dataset and BERT models](https://mlsanity.com/paper/2203.13778/) (2022)
- [L3Cube-MahaNews: News-based Short Text and Long Document Classification Datasets in Marathi](https://mlsanity.com/paper/2404.18216/) (2024)
- [Efficacy of BERT embeddings on predicting disaster from Twitter data](https://mlsanity.com/paper/2108.10698/) (2021)
- [Hate speech detection using static BERT embeddings](https://mlsanity.com/paper/2106.15537/) (2021)
- [Enhancing Depressive Post Detection in Bangla: A Comparative Study of TF-IDF, BERT and FastText Embeddings](https://mlsanity.com/paper/2407.09187/) (2024)

## Abstract (arXiv)

Natural Language Processing (NLP) for low-resource languages, which lack large annotated datasets, faces significant challenges due to limited high-quality data and linguistic resources. The selection of embeddings plays a critical role in achieving strong performance in NLP tasks. While contextual BERT embeddings require a full forward pass, non-contextual BERT embeddings rely only on table lookup. Existing research has primarily focused on contextual BERT embeddings, leaving non-contextual embeddings largely unexplored. In this study, we analyze the effectiveness of non-contextual embeddings from BERT models (MuRIL and MahaBERT) and FastText models (IndicFT and MahaFT) for tasks such as news classification, sentiment analysis, and hate speech detection in one such low-resource language Marathi. We compare these embeddings with their contextual and compressed variants. Our findings indicate that non-contextual BERT embeddings extracted from the model's first embedding layer outperform FastText embeddings, presenting a promising alternative for low-resource NLP.