Impact of Tokenization on Language Models: An Analysis for Turkish

no code implementations • 19 Apr 2022 • Cagri Toraman, Eyup Halit Yilmaz, Furkan Şahinuç, Oguzhan Ozcelik

Furthermore, we find that increasing the vocabulary size improves the performance of Morphological and Word-level tokenizers more than that of de facto tokenizers.

Large-Scale Hate Speech Detection with Cross-Domain Transfer

1 code implementation • LREC 2022 • Cagri Toraman, Furkan Şahinuç, Eyup Halit Yilmaz

The experimental results supported by statistical tests show that Transformer-based language models outperform conventional bag-of-words and neural models by at least 5% in English and 10% in Turkish for large-scale hate speech detection.

Hate Speech Detection Transfer Learning

ConQX: Semantic Expansion of Spoken Queries for Intent Detection based on Conditioned Text Generation

no code implementations • 2 Sep 2021 • Eyup Halit Yilmaz, Cagri Toraman

To provide additional information regarding the query and enhance the performance of intent detection, we propose a method for semantic expansion of spoken queries, called ConQX, which utilizes the text generation ability of an auto-regressive language model, GPT-2.

Few-Shot Learning Intent Detection +2