Counterfactual Debiasing for Generating Factually Consistent Text Summaries
This addresses the problem of generating factually consistent summaries for users relying on automated summarization systems, representing an incremental improvement over existing methods.
The paper tackles factual inconsistency in abstractive text summarization by proposing CoFactSum, a debiasing framework that uses counterfactual estimation to address language and irrelevancy biases, resulting in improved factual consistency on two datasets compared to baselines.
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.