MERMAID: Metaphor Generation with Symbolism and Discriminative Decoding
This addresses the problem of automated metaphor generation for natural language processing applications, representing an incremental improvement with specific gains in quality.
The paper tackles the challenge of generating metaphors by replacing verbs in literal expressions, using a method to automatically construct a parallel corpus from poetry and incorporating a discriminator to guide decoding. The best model generates metaphors better than baselines 66% of the time and enhances poems with metaphors preferred 68% of the time.
Generating metaphors is a challenging task as it requires a proper understanding of abstract concepts, making connections between unrelated concepts, and deviating from the literal meaning. In this paper, we aim to generate a metaphoric sentence given a literal expression by replacing relevant verbs. Based on a theoretically-grounded connection between metaphors and symbols, we propose a method to automatically construct a parallel corpus by transforming a large number of metaphorical sentences from the Gutenberg Poetry corpus (Jacobs, 2018) to their literal counterpart using recent advances in masked language modeling coupled with commonsense inference. For the generation task, we incorporate a metaphor discriminator to guide the decoding of a sequence to sequence model fine-tuned on our parallel data to generate high-quality metaphors. Human evaluation on an independent test set of literal statements shows that our best model generates metaphors better than three well-crafted baselines 66% of the time on average. A task-based evaluation shows that human-written poems enhanced with metaphors proposed by our model are preferred 68% of the time compared to poems without metaphors.