Interpretability

2D-OOB: Attributing Data Contribution Through Joint Valuation Framework NeurIPS 2024 Accelerating the Global Aggregation of Local Explanations AAAI 2024 A General Theoretical Framework for Learning Smallest Interpretable Models AAAI 2024 Almost-Linear RNNs Yield Highly Interpretable Symbolic Codes in Dynamical Systems Reconstruction NeurIPS 2024 An L* Algorithm for Deterministic Weighted Regular Languages EMNLP 2024 Are self-explanations from Large Language Models faithful? ACL 2024 Backward Lens: Projecting Language Model Gradients into the Vocabulary Space EMNLP 2024 Beyond Accuracy: Ensuring Correct Predictions With Correct Rationales NeurIPS 2024 Beyond Concept Bottleneck Models: How to Make Black Boxes Intervenable? NeurIPS 2024 Beyond Correlation: Interpretable Evaluation of Machine Translation Metrics EMNLP 2024 Beyond Label Attention: Transparency in Language Models for Automated Medical Coding via Dictionary Learning EMNLP 2024 BiasWipe: Mitigating Unintended Bias in Text Classifiers through Model Interpretability EMNLP 2024 Brain Decodes Deep Nets CVPR 2024 CAVA: A Tool for Cultural Alignment Visualization & Analysis EMNLP 2024 Cluster-Norm for Unsupervised Probing of Knowledge EMNLP 2024 CoSy: Evaluating Textual Explanations of Neurons NeurIPS 2024 Data-faithful Feature Attribution: Mitigating Unobservable Confounders via Instrumental Variables NeurIPS 2024 Decompose-and-Compose: A Compositional Approach to Mitigating Spurious Correlation CVPR 2024 Discovering Biases in Information Retrieval Models Using Relevance Thesaurus as Global Explanation EMNLP 2024 Disentangling Interpretable Factors with Supervised Independent Subspace Principal Component Analysis NeurIPS 2024 Dissect Black Box: Interpreting for Rule-Based Explanations in Unsupervised Anomaly Detection NeurIPS 2024 Dissecting Query-Key Interaction in Vision Transformers NeurIPS 2024 Does Large Language Model Contain Task-Specific Neurons? EMNLP 2024 Do Large Language Models Know How Much They Know? EMNLP 2024 Empowering CAM-Based Methods with Capability to Generate Fine-Grained and High-Faithfulness Explanations AAAI 2024