4 papers · 1 filter
Collaboratively adding new knowledge to an LLM
Rhui Dih Lee, Laura Wynter
We address the question of how to successively add new knowledge to an LLM whilst retaining previously-added knowledge. We consider two settings, semi-cooperative and fully-coopera…
Enhancing Training Efficiency Using Packing with Flash Attention
Achintya Kundu, Rhui Dih Lee, Laura Wynter +2
Padding is often used in tuning LLM models by adding special tokens to shorter training examples to match the length of the longest sequence in each batch. While this ensures unifo…
Efficiently Distilling LLMs for Edge Applications
Achintya Kundu, Fabian Lim, Aaron Chew +3
Supernet training of LLMs is of great interest in industrial applications as it confers the ability to produce a palette of smaller models at constant cost, regardless of the numbe…
Transfer-Once-For-All: AI Model Optimization for Edge
Achintya Kundu, Laura Wynter, Rhui Dih Lee +1
Weight-sharing neural architecture search aims to optimize a configurable neural network model (supernet) for a variety of deployment scenarios across many devices with different r…