TY - GEN

T1 - Rather a Nurse than a Physician - Contrastive Explanations under Investigation

AU - Eberle, Oliver

AU - Chalkidis, Ilias

AU - Cabello, Laura

AU - Brandl, Stephanie

PY - 2023

Y1 - 2023

N2 - Contrastive explanations, where one decision is explained *in contrast to another*, are supposed to be closer to how humans explain a decision than non-contrastive explanations, where the decision is not necessarily referenced to an alternative. This claim has never been empirically validated. We analyze four English text-classification datasets (SST2, DynaSent, BIOS and DBpedia-Animals). We fine-tune and extract explanations from three different models (RoBERTa, GTP-2, and T5), each in three different sizes and apply three post-hoc explainability methods (LRP, GradientxInput, GradNorm). We furthermore collect and release human rationale annotations for a subset of 100 samples from the BIOS dataset for contrastive and non-contrastive settings. A cross-comparison between model-based rationales and human annotations, both in contrastive and non-contrastive settings, yields a high agreement between the two settings for models as well as for humans. Moreover, model-based explanations computed in both settings align equally well with human rationales. Thus, we empirically find that humans do not necessarily explain in a contrastive manner.

AB - Contrastive explanations, where one decision is explained *in contrast to another*, are supposed to be closer to how humans explain a decision than non-contrastive explanations, where the decision is not necessarily referenced to an alternative. This claim has never been empirically validated. We analyze four English text-classification datasets (SST2, DynaSent, BIOS and DBpedia-Animals). We fine-tune and extract explanations from three different models (RoBERTa, GTP-2, and T5), each in three different sizes and apply three post-hoc explainability methods (LRP, GradientxInput, GradNorm). We furthermore collect and release human rationale annotations for a subset of 100 samples from the BIOS dataset for contrastive and non-contrastive settings. A cross-comparison between model-based rationales and human annotations, both in contrastive and non-contrastive settings, yields a high agreement between the two settings for models as well as for humans. Moreover, model-based explanations computed in both settings align equally well with human rationales. Thus, we empirically find that humans do not necessarily explain in a contrastive manner.

U2 - 10.18653/v1/2023.emnlp-main.427

DO - 10.18653/v1/2023.emnlp-main.427

M3 - Article in proceedings

SP - 6907

EP - 6920

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 -