Quanda: An Interpretability Toolkit for Training Data Attribution Evaluation and Beyond
Not on the map. Why?
Abstract (arXiv)
In recent years, training data attribution (TDA) methods have emerged as a promising direction for the interpretability of neural networks. While research around TDA is thriving, limited effort has been dedicated to the evaluation of attributions. Similar to the development of evaluation metrics for traditional feature attribution approaches, several standalone metrics have been proposed to evaluate the quality of TDA methods across various contexts. However, the lack of a unified framework that allows for systematic comparison limits trust in TDA methods and stunts their widespread adoption. To address this research gap, we introduce Quanda, a Python toolkit designed to facilitate the evaluation of TDA methods. Beyond offering a comprehensive set of evaluation metrics, Quanda provides a uniform interface for seamless integration with existing TDA implementations across different repositories, thus enabling systematic benchmarking. The toolkit is user-friendly, thoroughly tested, well-documented, and available as an open-source library on PyPi and under https://github.com/dilyabareeva/quanda.
Key facts. Cited by 2 papers in the map (data through 14 Oct 2026). Builds on 29 of its 59 references that are in the map. First posted on arXiv 9 Oct 2024.
Builds on
29 of 59 references.
References · 29
- 2019PyTorch: An Imperative Style, High-Performance Deep Learning Librarycited by 20,305
- 2017A Unified Approach to Interpreting Model Predictionscited by 6,555
- 2017Axiomatic Attribution for Deep Networkscited by 3,384
- 2018Neural Tangent Kernel: Convergence and Generalization in Neural Networkscited by 2,621
- 2013Deep Inside Convolutional Networks: Visualising Image Classification Models and Saliency Mapscited by 2,371
- 2017Decoupled Weight Decay Regularizationcited by 1,831
- 2017Understanding Black-box Predictions via Influence Functionscited by 1,741
- 2019Learning Robust Global Representations by Penalizing Local Predictive Powercited by 1,013
- 2020Estimating Training Data Influence by Tracing Gradient Descentcited by 580
- 2020Captum: A unified and generic model interpretability library for PyTorchcited by 547
- 2021AI and the Everything in the Whole Wide World Benchmarkcited by 349
- 2023TRAK: Attributing Model Behavior at Scalecited by 283
17 more
- 2020Influence Functions in Deep Learning Are Fragilecited by 252
- 2019InterpretML: A Unified Framework for Machine Learning Interpretabilitycited by 246
- 2022Quantus: An Explainable AI Toolkit for Responsible Evaluation of Neural Network Explanations and Beyondcited by 171
- 2022Dataset Pruning: Reducing Training Data by Examining Generalization Influencecited by 133
- 2022OpenXAI: Towards a Transparent Evaluation of Model Explanationscited by 126
- 2023DataInf: Efficiently Estimating Data Influence in LoRA-tuned LLMs and Diffusion Modelscited by 123
- 2024DsDm: Model-Aware Dataset Selection with Datamodelscited by 105
- 2023Concept-based Explainable Artificial Intelligence: A Surveycited by 98
- 2021Input Similarity from the Neural Network Perspectivecited by 79
- 2021Revisiting Methods for Finding Influential Examplescited by 39
- 2024Training Data Attribution via Approximate Unrolled Differentiationcited by 34
- 2022Xplique: A Deep Learning Explainability Toolboxcited by 24
- 2023When Less is More: Investigating Data Pruning for Pretraining LLMs at Scalecited by 10
- 2024Sparse, Efficient and Explainable Data Attribution with DualXDAcited by 6
- 2023A Simple and Efficient Baseline for Data Attribution on Imagescited by 5
- 2023Exploring Practitioner Perspectives On Training Data Attribution Explanationscited by 3
- 2020Evaluation of Similarity-based Explanationscited by 2
Built on by
Citing papers per year: 2025: 2.
2 papers.