Paper

AdvSumm: Adversarial Training for Bias Mitigation in Text Summarization

Subbiah, Melanie, McKeown, Kathleen, Nikhil Reddy Varimalla, Gupta, Mukur, Deas, Nicholas

2025

Overview

Large Language Models (LLMs) have achieved impressive performance in text summarization and are increasingly deployed in real-world applications. However, these systems often inherit associative and framing biases from pre-training data, leading to inappropriate or unfair outputs in downstream tasks. In this work, we present AdvSumm (Adversarial Summarization), a domain-agnostic training framework designed to mitigate bias in text summarization through improved generalization. Inspired by adversarial robustness, AdvSumm introduces a novel Perturber component that applies gradient-guided perturbations at the embedding level of Sequence-to-Sequence models, enhancing the model's robustness to input variations. We empirically demonstrate that AdvSumm effectively reduces different types of bias in summarization-specifically, name-nationality bias and political framing bias-without compromising summarization quality. Compared to standard transformers and data augmentation techniques like back-translation, AdvSumm achieves stronger bias mitigation performance across benchmark datasets.

• Publisher: Cornell University Library, arXiv.org
• Subject: Bias / Data augmentation / Large language models / Robustness