model debugging

HiBug2: Efficient and Interpretable Error Slice Discovery for Comprehensive Model Debugging ICLR 2025 TAB: Transformer Attention Bottlenecks enable User Intervention and Debugging in Vision-Language Models ICCV 2025 Training with Confidence: Catching Silent Errors in Deep Learning Training with Automated Proactive Checks OSDI 2025 Beyond Accuracy: Evaluating Self-Consistency of Code Large Language Models with IdentityChain ICLR 2024 Token-wise Influential Training Data Retrieval for Large Language Models ACL 2024 Class based Influence Functions for Error Detection ACL 2023 Error Discovery By Clustering Influence Embeddings NeurIPS 2023 HiBug: On Human-Interpretable Model Debug NeurIPS 2023 IAEval: A Comprehensive Evaluation of Instance Attribution on Natural Language Understanding EMNLP 2023 Red Teaming Deep Neural Networks with Feature Synthesis Tools NeurIPS 2023 Theoretical and Practical Perspectives on what Influence Functions Do NeurIPS 2023 3DB: A Framework for Debugging Computer Vision Models NeurIPS 2022 Debugging and Explaining Metric Learning Approaches: An Influence Function Based Perspective NeurIPS 2022 Meaningfully debugging model mistakes using conceptual counterfactual explanations ICML 2022 Missingness Bias in Model Debugging ICLR 2022 Model Doctor: A Simple Gradient Aggregation Strategy for Diagnosing and Treating CNN Classifiers AAAI 2022 Robust Feature-Level Adversaries are Interpretability Tools NeurIPS 2022 Use-Case-Grounded Simulations for Explanation Evaluation NeurIPS 2022 “Will You Find These Shortcuts?” A Protocol for Evaluating the Faithfulness of Input Salience Methods for Text Classification EMNLP 2022 An Empirical Comparison of Instance Attribution Methods for NLP NAACL 2021 Leveraging Sparse Linear Layers for Debuggable Deep Networks ICML 2021 Towards Benchmarking the Utility of Explanations for Model Debugging NAACL 2021 Understanding Failures of Deep Networks via Robust Feature Extraction CVPR 2021 Debugging Tests for Model Explanations NeurIPS 2020 Debugging Sequence-to-Sequence Models with Seq2Seq-Vis EMNLP 2018