conftrace_

padding-free training

1 paper

Explore in graph

Co-occurring keywords · papers mentioning both

bucket sampling (1) padding token (1) efficiency optimization (1) sequence padding (1) token dropping (1) model regularization (1) overfitting prevention (1) batch sampling (1) model efficiency (1) efficient fine-tuning (1)

Papers

TokenDrop + BucketSampler: Towards Efficient Padding-free Fine-tuning of Language Models EMNLP 2023