CONFIT: Toward Faithful Dialogue Summarization with Linguistically-Informed Contrastive Fine-tuning

Understand

Factual inconsistencies in generated summaries severely limit the practical applications of abstractive dialogue summarization.

• Although significant progress has been achieved by using pre-trained models, substantial amounts of hallucinated content are found during the human evaluation.
• Pre-trained models are most commonly fine-tuned with cross-entropy loss for text summarization, which may not be an optimal strategy.
• In this work, we provide a typology of factual errors with annotation data to highlight the types of errors and move away from a binary understanding of factuality.