Investigating the Pre-Training Dynamics of In-Context Learning: Task Recognition vs. Task Learning
This work addresses the understanding of in-context learning emergence for AI researchers, offering insights into managing competitive abilities to improve efficiency, though it is incremental as it builds on existing ICL frameworks.
The paper investigates the pre-training dynamics of in-context learning, finding that task recognition and task learning abilities are competitive and negatively correlated with ICL performance, and proposes an adaptive ensemble method that boosts ICL performance, enabling two small models to outperform a larger one with over twice the parameters.
The emergence of in-context learning (ICL) is potentially attributed to two major abilities: task recognition (TR) for recognizing the task from demonstrations and utilizing pre-trained priors, and task learning (TL) for learning from demonstrations. However, relationships between the two abilities and how such relationships affect the emergence of ICL is unclear. In this paper, we take the first step by examining the pre-training dynamics of the emergence of ICL. With carefully designed metrics, we find that these two abilities are, in fact, competitive during pre-training. Moreover, we observe a strong negative correlation between the competition and ICL performance. Further analysis of common pre-training factors (i.e., model size, dataset size, and data curriculum) demonstrates possible ways to manage the competition. Based on these insights, we propose a simple yet effective method to better integrate these two abilities for ICL at inference time. Through adaptive ensemble learning, the performance of ICL can be significantly boosted, enabling two small models to outperform a larger one with more than twice the parameters. The code is available at https://github.com/RUCAIBox/Competitive-ICL.