paper-with-me

홈 › Papers

Counterfactual Debiasing for Generating Factually Consistent Text Summaries

2023-05-18 · Chenhe Dong, Yuexiang Xie, Yaliang Li, Ying Shen

Despite substantial progress in abstractive text summarization to generate fluent and informative texts, the factual inconsistency in the generated summaries remains an important yet challenging problem to be solved. In this paper, we construct causal graphs for abstractive text summarization and identify the intrinsic causes of the factual inconsistency, i.e., the language bias and irrelevancy bias, and further propose a debiasing framework, named CoFactSum, to alleviate the causal effects of these biases by counterfactual estimation. Specifically, the proposed CoFactSum provides two counterfactual estimation strategies, i.e., Explicit Counterfactual Masking with an explicit dynamic masking strategy, and Implicit Counterfactual Training with an implicit discriminative cross-attention mechanism. Meanwhile, we design a Debiasing Degree Adjustment mechanism to dynamically adapt the debiasing degree at each decoding step. Extensive experiments on two widely-used summarization datasets demonstrate the effectiveness of CoFactSum in enhancing the factual consistency of generated summaries compared with several baselines.

📄 PDF Abstract BibTeX arXiv:2305.10736

Code (0)

등록된 구현이 없습니다.

Tasks

Abstractive Text SummarizationcounterfactualText Summarization

Similar Papers 제목 키워드 기반

CoLD: Counterfactually-Guided Length Debiasing for Process Reward Models in Mathematical Reasoning

2025-07-21 · Congmin Zheng, Jiachen Zhu, Jianghao Lin, Xinyi Dai 외 arxiv

Process Reward Models (PRMs) play a central role in evaluating and guiding multi-step reasoning in large language models (LLMs), especially for mathematical problem solving. However, we identify a pervasive length bias i…

Mathematical Reasoning

Debiasing Text Safety Classifiers through a Fairness-Aware Ensemble

2024-09-05 · Olivia Sturman, Aparna Joshi, Bhaktipriya Radharapu, Piyush Kumar 외

Increasing use of large language models (LLMs) demand performant guardrails to ensure the safety of inputs and outputs of LLMs. When these safeguards are trained on imbalanced data, they can learn the societal biases. We…

counterfactualFairness

Counterfactually Measuring and Eliminating Social Bias in Vision-Language Pre-training Models

2022-07-03 · Yi Zhang, Junyang Wang, Jitao Sang

Vision-Language Pre-training (VLP) models have achieved state-of-the-art performance in numerous cross-modal tasks. Since they are optimized to capture the statistical properties of intra- and inter-modality, there remai…

counterfactual

All Should Be Equal in the Eyes of Language Models: Counterfactually Aware Fair Text Generation

2023-11-09 · Pragyan Banerjee, Abhinav Java, Surgan Jandial, Simra Shahid 외

Fairness in Language Models (LMs) remains a longstanding challenge, given the inherent biases in training data that can be perpetuated by models and affect the downstream tasks. Recent methods employ expensive retraining…

AllFairnessLanguage ModelingLanguage Modelling+1

MABEL: Attenuating Gender Bias using Textual Entailment Data

2022-10-26 · Jacqueline He, Mengzhou Xia, Christiane Fellbaum, Danqi Chen

Pre-trained language models encode undesirable social biases, which are further exacerbated in downstream use. To this end, we propose MABEL (a Method for Attenuating Gender Bias using Entailment Labels), an intermediate…

Contrastive LearningFairnessNatural Language Inference