Quantifying Adaptability in Pre-trained Language Models with 500 Tasks. (arXiv:2112.03204v2 [cs.CL] UPDATED)

Web: http://arxiv.org/abs/2112.03204

When a neural language model (LM) is adapted to perform a new task, what
aspects of the task predict the eventual performance of the model? In NLP,
systematic features of LM generalization to individual examples are well
characterized, but systematic aspects of LM adaptability to new tasks are not
nearly as well understood. We present a large-scale empirical study of the
features and limits of LM adaptability using a new benchmark, TaskBench500,
built from 500 procedurally generated sequence modeling tasks. …

arxiv language language models models