Mention Flags (MF): Constraining Transformer-based Text Generators

This paper focuses on Seq2Seq (S2S) constrained text generation where the text generator is constrained to mention specific words, which are inputs to the encoder, in the generated outputs. Pre-trained S2S models such as T5 or a Copy Mechanism can be trained to copy the surface tokens from encoders to decoders, but they cannot guarantee constraint satisfaction. Constrained decoding algorithms always produce hypotheses satisfying all constraints. However, they are computationally expensive and can lower the generated text quality. In this paper, we propose Mention Flags (MF), which trace whether lexical constraints are satisfied in the generated outputs of an S2S decoder. The MF models are trained to generate tokens until all constraints are satisfied, guaranteeing high constraint satisfaction. Our experiments on the Common Sense Generation task (CommonGen) (Lin et al., 2020), End2end Data-to-Text task (E2ENLG) (Dušek et al., 2020) and Novel Object Captioning task (nocaps) (Agrawal et al., 2019) show that the MF models maintain higher constraint satisfaction and text quality than the baseline models and other constrained text generation algorithms, achieving state-of-the-art performance on all three tasks. These results are achieved with a much lower run-time than constrained decoding algorithms. We also show that the MF models work well in the low-resource setting. 1