Learning or Self-aligning? Rethinking Instruction Fine-tuning
This work addresses a fundamental problem in AI for researchers and practitioners by revealing that IFT may not effectively transfer knowledge, which is incremental but challenges common assumptions.
The paper investigates the mechanisms of Instruction Fine-Tuning (IFT) in large language models, finding that attempts to learn additional world knowledge through IFT often fail or have negative effects, and that maintaining internal knowledge consistency is critical for success.
Instruction Fine-tuning~(IFT) is a critical phase in building large language models~(LLMs). Previous works mainly focus on the IFT's role in the transfer of behavioral norms and the learning of additional world knowledge. However, the understanding of the underlying mechanisms of IFT remains significantly limited. In this paper, we design a knowledge intervention framework to decouple the potential underlying factors of IFT, thereby enabling individual analysis of different factors. Surprisingly, our experiments reveal that attempting to learn additional world knowledge through IFT often struggles to yield positive impacts and can even lead to markedly negative effects. Further, we discover that maintaining internal knowledge consistency before and after IFT is a critical factor for achieving successful IFT. Our findings reveal the underlying mechanisms of IFT and provide robust support for some very recent and potential future works.