Mitigating Hallucination in Abstractive Summarization with Domain-Conditional Mutual Information

arXiv:2404.09480v1 Announce Type: new
Abstract: A primary challenge in abstractive summarization is hallucination -- the phenomenon where a model generates plausible text that is absent in the source text. We hypothesize that the domain (or topic) of the source text triggers the model to generate text that is highly probable in the domain, neglecting the details of the source text. To alleviate this model bias, we introduce a decoding strategy based on domain-conditional pointwise mutual information. This strategy adjusts the …

arxiv cs.ai cs.cl domain hallucination information summarization type