MIT Press

Table 1:

Story examples generated by human and GPT-2 models. The stories written by the pretrained GPT-2 and fine-tuned GPT-2 (post-trained on ROCStories [Mostafazadeh et al., 2016b]) suffer from repetition (in italic), bad inter-sentence coherence to the context (e.g., ignoring key entities such as accident in bold), as well as conflicting logic (underlined, e.g., first stopped driving and then drove to the party), in spite of their good fluency and intra-sentence coherence.

Context:The first time I saw the results of an accident, I was in a bus.

Golden Continuation by Human:We were stopped and a friend pointed out the accident on the road. A man was under a car after having been run over. People tried to move the car off him but only made it worse. After seeing that, I am much more careful when crossing the street.

pretrained GPT-2:She was trying to get home but the other passengers would not let her go. He thought she was going to die because of her weight. She was trying to get home but the other passengers would not let her go. The bus driver would not let her go.

Fine-tuned GPT-2:I was on my way to a party. I was on my way to a party and I ’d gotten out of my seat, and started driving. I got a flat tire, so I stopped driving. I drove to the party and had a great time.

View Large

This Feature Is Available To Subscribers Only