Reasoning or Reciting? Exploring the Capabilities and Limitations of Language Models Through Counterfactual Tasks

arXiv:2307.02477v3 Announce Type: replace
Abstract: The impressive performance of recent language models across a wide range of tasks suggests that they possess a degree of abstract reasoning skills. Are these skills general and transferable, or specialized to specific tasks seen during pretraining? To disentangle these effects, we propose an evaluation framework based on "counterfactual" task variants that deviate from the default assumptions underlying standard tasks. Across a suite of 11 tasks, we observe nontrivial performance on the counterfactual variants, but …

abstract arxiv capabilities counterfactual cs.ai cs.cl effects general language language models limitations performance pretraining reasoning skills specific tasks tasks through type