Niloy Ganguly

Fine-grained Sentiment Controlled Text Generation

2020 · arXiv (Cornell University)

Controlled text generation techniques aim to regulate specific attributes (e.g. sentiment) while preserving the attribute independent content. The state-of-the-art approaches model the specified attribute as a structured or discrete representation while making the content representation …

Adaptive BPE Tokenization for Enhanced Vocabulary Adaptation in Finetuning Pretrained Language Models

2024

In this work, we show a fundamental limitation in vocabulary adaptation approaches that use Byte-Pair Encoding (BPE) tokenization scheme for fine-tuning pretrained language models (PLMs) to expert domains.Current approaches trivially append the target domainspecific vocabulary …