Why Exposure Bias Matters: An Imitation Learning Perspective of Error Accumulation in Language Generation. (arXiv:2204.01171v2 [cs.CL] UPDATED)

Web: http://arxiv.org/abs/2204.01171

Current language generation models suffer from issues such as repetition,
incoherence, and hallucinations. An often-repeated hypothesis is that this
brittleness of generation models is caused by the training and the generation
procedure mismatch, also referred to as exposure bias. In this paper, we verify
this hypothesis by analyzing exposure bias from an imitation learning
perspective. We show that exposure bias leads to an accumulation of errors,
analyze why perplexity fails to capture this accumulation, and empirically show
that this accumulation …

arxiv bias imitation learning language language generation learning perspective