Improving Factual Consistency of News Summarization by Contrastive Preference Optimization
Abstract: Despite the recent progress in news summarization made by large language models (LLMs), they often generate summaries that are factually inconsistent with original articles, known as "hallucinations" in text generation. Unlike previous small models (e.g., BART, T5), current LLMs make fewer silly mistakes but more sophisticated ones, such as imposing cause and effect, adding false details, overgeneralizing, etc. These hallucinations are challenging to detect through traditional methods, which poses great challenges for improving the factual consistency of text summarization. In this paper, we propose **C**ontrastive **P**reference **O**ptimization (**CPO**) to disentangle the LLMs' propensities to generate faithful and fake content. Furthermore, we adopt a probing-based specific training method to improve their capacity of distinguishing two types of generation. In this way, LLMs can execute the instructions more accurately and have enhanced perception of hallucinations. Experimental results show that CPO significantly improves the reliability of summarization based on LLMs.
Paper Type: Long
Research Area: Summarization
Research Area Keywords: Summarization, Hallucinations, Probing
Contribution Types: Model analysis & interpretability, NLP engineering experiment, Data resources
Languages Studied: English
Submission Number: 1072
Loading