Improving Instruct Models for Free: A Study on Partial Adaptation
This work addresses the problem of performance degradation in instruct models for researchers and practitioners, offering an incremental method to optimize trade-offs in model adaptation.
The study investigates the trade-off between instruction following and in-context learning in instruct models, finding that reducing instruction-tuning strength improves few-shot performance on natural language tasks by up to 15% across model families, at the cost of some instruction following ability.
Instruct models, obtained from various instruction tuning or post-training steps, are commonly deemed superior and more usable than their base counterpart. While the model gains instruction following ability, instruction tuning may lead to forgetting the knowledge from pre-training or it may encourage the model to become overly conversational or verbose. This, in turn, can lead to degradation of in-context few-shot learning performance. In this work, we study the performance trajectory between base and instruct models by scaling down the strength of instruction-tuning via the partial adaption method. We show that, across several model families and model sizes, reducing the strength of instruction-tuning results in material improvement on a few-shot in-context learning benchmark covering a variety of classic natural language tasks. This comes at the cost of losing some degree of instruction following ability as measured by AlpacaEval. Our study shines light on the potential trade-off between in-context learning and instruction following abilities that is worth considering in practice.