In-Context Learning with Long-Context Models: An In-Depth Exploration

arXiv:2405.00200v1 Announce Type: new
Abstract: As model context lengths continue to increase, the number of demonstrations that can be provided in-context approaches the size of entire training datasets. We study the behavior of in-context learning (ICL) at this extreme scale on multiple datasets and models. We show that, for many datasets with large label spaces, performance continues to increase with hundreds or thousands of demonstrations. We contrast this with example retrieval and finetuning: example retrieval shows excellent performance at low …

abstract arxiv behavior context cs.cl datasets exploration in-context learning multiple scale show study training training datasets type