A comparison of svm against pre-trained language models (plms) for text classification tasks

Y Wahba, N Madhavji, J Steinbacher - International Conference on …, 2022 - Springer
The emergence of pre-trained language models (PLMs) has shown great success in many
Natural Language Processing (NLP) tasks including text classification. Due to the minimal to …

Less is more: Pruning BERTweet architecture in Twitter sentiment analysis

R Moura, J Carvalho, A Plastino, A Paes - Information Processing & …, 2024 - Elsevier
Transformer-based models have been scaled up to account for absorbing more information
and improve their performances. However, several studies have called attention to their …

Tuning Language Models by Mixture-of-Depths Ensemble

H Luo, L Specia - arxiv preprint arxiv:2410.13077, 2024 - arxiv.org
Transformer-based Large Language Models (LLMs) traditionally rely on final-layer loss for
training and final-layer representations for predictions, potentially overlooking the predictive …

Mitigating Hallucination Issues in Small-Parameter LLMs through Inter-Layer Contrastive Decoding

F Li, P Zhang - … Joint Conference on Neural Networks (IJCNN …, 2024 - ieeexplore.ieee.org
In this paper, we introduce a new decoding method to mitigate the issue of hallucinations in
Large Language Models (LLMs). Specifically, our method dynamically selects appropriate …