Interpretability

Fair and Interpretable Algorithmic Hiring using Evolutionary Many Objective Optimization AAAI 2021 Fair Comparison: Quantifying Variance in Results for Fine-Grained Visual Categorization WACV 2021 Faithfully Explainable Recommendation via Neural Logic Reasoning NAACL 2021 Farewell to Mutual Information: Variational Distillation for Cross-Modal Person Re-Identification CVPR 2021 Fast Axiomatic Attribution for Neural Networks NeurIPS 2021 FastIF: Scalable Influence Functions for Efficient Model Interpretation and Debugging EMNLP 2021 Feature Space Targeted Attacks by Statistic Alignment IJCAI 2021 FIMAP: Feature Importance by Minimal Adversarial Perturbation AAAI 2021 Finding Discriminative Filters for Specific Degradations in Blind Super-Resolution NeurIPS 2021 Finding Representative Interpretations on Convolutional Neural Networks ICCV 2021 Fine-grained Factual Consistency Assessment for Abstractive Summarization Models EMNLP 2021 Fine-grained Interpretation and Causation Analysis in Deep NLP Models NAACL 2021 Fine-Grained Neural Network Explanation by Identifying Input Features with Predictive Information NeurIPS 2021 Fine-Tuned Transformers Show Clusters of Similar Representations Across Layers EMNLP 2021 FITAnnotator: A Flexible and Intelligent Text Annotation System NAACL 2021 Flow-based Attribution in Graphical Models: A Recursive Shapley Approach ICML 2021 Focus Attention: Promoting Faithfulness and Diversity in Summarization ACL 2021 Foundations of Symbolic Languages for Model Interpretability NeurIPS 2021 FrameNet and Typology NAACL 2021 From global to local MDI variable importances for random forests and when they are Shapley values NeurIPS 2021 From the Stage to the Audience: Propaganda on Reddit EACL 2021 Gaining Insight into SARS-CoV-2 Infection and COVID-19 Severity Using Self-supervised Edge Features and Graph Neural Networks AAAI 2021 GANMEX: One-vs-One Attributions using GAN-based Model Explainability ICML 2021 Generalizable Multi-linear Attention Network NeurIPS 2021 Generalization in NLI: Ways (Not) To Go Beyond Simple Heuristics EMNLP 2021