Nutribullets Hybrid: Multi-document Health Summarization
arXiv:2104.03465
Abstract
We present a method for generating comparative summaries that highlights similarities and contradictions in input documents. The key challenge in creating such summaries is the lack of large parallel training data required for training typical summarization systems. To this end, we introduce a hybrid generation approach inspired by traditional concept-to-text systems. To enable accurate comparison between different sources, the model first learns to extract pertinent relations from input documents. The content planning component uses deterministic operators to aggregate these relations after identifying a subset for inclusion into a summary. The surface realization component lexicalizes this information using a text-infilling language model. By separately modeling content selection and realization, we can effectively train them with limited annotations. We implemented and tested the model in the domain of nutrition and health -- rife with inconsistencies. Compared to conventional methods, our framework leads to more faithful, relevant and aggregation-sensitive summarization -- while being equally fluent.
NAACL 2021 Camera Ready
References in corpus (5)
- Insertion Transformer: Flexible Sequence Generation via Insertion Operations
- Query Focused Abstractive Summarization: Incorporating Query Relevance, Multi-Document Coverage, and Summary Length Constraints into seq2seq Models
- Efficient Adaptation of Pretrained Transformers for Abstractive Summarization
- Enabling Language Models to Fill in the Blanks
- Concept Pointer Network for Abstractive Summarization