Bias in News Summarization: Measures, Pitfalls and Corpora
This addresses the problem of harmful social biases in summarization outputs for researchers and practitioners, though it is incremental in building on known bias issues in LLMs.
The paper tackles the problem of measuring bias in news summarization by LLMs, introducing definitions and a method to generate controlled input documents, and finds that while content selection is largely unaffected by gender bias, hallucinations show evidence of bias.
Summarization is an important application of large language models (LLMs). Most previous evaluation of summarization models has focused on their content selection, faithfulness, grammaticality and coherence. However, it is well known that LLMs can reproduce and reinforce harmful social biases. This raises the question: Do biases affect model outputs in a constrained setting like summarization? To help answer this question, we first motivate and introduce a number of definitions for biased behaviours in summarization models, along with practical operationalizations. Since we find that biases inherent to input documents can confound bias analysis in summaries, we propose a method to generate input documents with carefully controlled demographic attributes. This allows us to study summarizer behavior in a controlled setting, while still working with realistic input documents. We measure gender bias in English summaries generated by both purpose-built summarization models and general purpose chat models as a case study. We find content selection in single document summarization to be largely unaffected by gender bias, while hallucinations exhibit evidence of bias. To demonstrate the generality of our approach, we additionally investigate racial bias, including intersectional settings.