CausalLM is not optimal for in-context learning
This work addresses a fundamental limitation in transformer-based in-context learning for machine learning practitioners, though it is incremental as it builds on known empirical observations.
The paper theoretically and empirically shows that prefix language models (prefixLM) outperform causal language models (causalLM) for in-context learning, with prefixLM converging to optimal linear regression solutions while causalLM behaves like suboptimal online gradient descent.
Recent empirical evidence indicates that transformer based in-context learning performs better when using a prefix language model (prefixLM), in which in-context samples can all attend to each other, compared to causal language models (causalLM), which use auto-regressive attention that prohibits in-context samples to attend to future samples. While this result is intuitive, it is not understood from a theoretical perspective. In this paper we take a theoretical approach and analyze the convergence behavior of prefixLM and causalLM under a certain parameter construction. Our analysis shows that both LM types converge to their stationary points at a linear rate, but that while prefixLM converges to the optimal solution of linear regression, causalLM convergence dynamics follows that of an online gradient descent algorithm, which is not guaranteed to be optimal even as the number of samples grows infinitely. We supplement our theoretical claims with empirical experiments over synthetic and real tasks and using various types of transformers. Our experiments verify that causalLM consistently underperforms prefixLM in all settings.