TY - GEN

T1 - Explaining Interactions Between Text Spans

AU - Choudhury, Sagnik

AU - Atanasova, Pepa

AU - Augenstein, Isabelle

PY - 2023

Y1 - 2023

N2 - Reasoning over spans of tokens from different parts of the input is essential for natural language understanding (NLU) tasks such as fact-checking (FC), machine reading comprehension (MRC) or natural language inference (NLI). However, existing highlight-based explanations primarily focus on identifying individual important features or interactions only between adjacent tokens or tuples of tokens. Most notably, there is a lack of annotations capturing the human decision-making process with respect to the necessary interactions for informed decision-making in such tasks. To bridge this gap, we introduce SpanEx, a multi-annotator dataset of human span interaction explanations for two NLU tasks: NLI and FC. We then investigate the decision-making processes of multiple fine-tuned large language models in terms of the employed connections between spans in separate parts of the input and compare them to the human reasoning processes. Finally, we present a novel community detection based unsupervised method to extract such interaction explanations. We make the code and the dataset available on [Github](https://github.com/copenlu/spanex). The dataset is also available on [Huggingface datasets](https://huggingface.co/datasets/copenlu/spanex).

AB - Reasoning over spans of tokens from different parts of the input is essential for natural language understanding (NLU) tasks such as fact-checking (FC), machine reading comprehension (MRC) or natural language inference (NLI). However, existing highlight-based explanations primarily focus on identifying individual important features or interactions only between adjacent tokens or tuples of tokens. Most notably, there is a lack of annotations capturing the human decision-making process with respect to the necessary interactions for informed decision-making in such tasks. To bridge this gap, we introduce SpanEx, a multi-annotator dataset of human span interaction explanations for two NLU tasks: NLI and FC. We then investigate the decision-making processes of multiple fine-tuned large language models in terms of the employed connections between spans in separate parts of the input and compare them to the human reasoning processes. Finally, we present a novel community detection based unsupervised method to extract such interaction explanations. We make the code and the dataset available on [Github](https://github.com/copenlu/spanex). The dataset is also available on [Huggingface datasets](https://huggingface.co/datasets/copenlu/spanex).

U2 - 10.18653/v1/2023.emnlp-main.783

DO - 10.18653/v1/2023.emnlp-main.783

M3 - Article in proceedings

SN - N 979-8-89176-060-8

SP - 12709

EP - 12730

BT - Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing

PB - Association for Computational Linguistics (ACL)

T2 - 2023 Conference on Empirical Methods in Natural Language Processing

Y2 - 6 December 2023 through 10 December 2023

ER -