{"id":14112690,"url":"https://github.com/maxidl/awesome-instance-attribution","last_synced_at":"2025-08-01T15:30:58.196Z","repository":{"id":129608091,"uuid":"481956856","full_name":"maxidl/awesome-instance-attribution","owner":"maxidl","description":"A collection of resources about instance attribution (in deep learning).","archived":false,"fork":false,"pushed_at":"2022-04-19T10:26:49.000Z","size":6,"stargazers_count":1,"open_issues_count":0,"forks_count":1,"subscribers_count":2,"default_branch":"main","last_synced_at":"2024-04-22T12:04:55.013Z","etag":null,"topics":[],"latest_commit_sha":null,"homepage":null,"language":null,"has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"mit","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/maxidl.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null}},"created_at":"2022-04-15T12:50:21.000Z","updated_at":"2022-04-25T12:03:47.000Z","dependencies_parsed_at":"2023-07-30T14:15:38.861Z","dependency_job_id":null,"html_url":"https://github.com/maxidl/awesome-instance-attribution","commit_stats":null,"previous_names":[],"tags_count":0,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/maxidl%2Fawesome-instance-attribution","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/maxidl%2Fawesome-instance-attribution/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/maxidl%2Fawesome-instance-attribution/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/maxidl%2Fawesome-instance-attribution/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/maxidl","download_url":"https://codeload.github.com/maxidl/awesome-instance-attribution/tar.gz/refs/heads/main","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":228389072,"owners_count":17912184,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":[],"created_at":"2024-08-14T10:03:52.926Z","updated_at":"2024-12-05T23:31:30.082Z","avatar_url":"https://github.com/maxidl.png","language":null,"readme":"# awesome-instance-attribution ![Awesome](figures/awesome.svg)\nA collection of resources on instance attribution (in deep learning).\n\nDid we miss a relevant resource? Feel free to raise an Issue or create a Pull Request.\n\n\n## Table of Contents\n- [ToDos](#todos)\n- [Methods](#methods)\n- [Evaluation](#evaluation)\n- [Other](#other)\n\n\n## [ToDos](#content)\n- Scout papers for additional resources\n- add github links for many papers\n- sort papers by some attribute? (e.g. publication data or arxiv date?)\n- write short gist for each resource. E.g. what research questions is it about? what are the key insights?\n\n## [Methods](#content)\n- **Influential Instances**\n    \n    Chapter in the Interpretable ML Book by Christoph Molnar.\n\n    [book chapter](https://christophm.github.io/interpretable-ml-book/influential.html)\n\n- **Understanding Black-box Predictions via Influence Functions**:\n\n    Koh and Liang. ICML 2017 (Best Paper)\n\n    [arxiv](https://arxiv.org/abs/1703.04730) [github](https://github.com/kohpangwei/influence-release)\n\n\n- **Efficient Estimation of Influence of a Training Instance**\n    \n    Kobayashi et al. SustaiNLP2020\n\n    [arxiv](https://arxiv.org/abs/2012.04207)\n\n- **FastIF: Scalable Influence Functions for Efficient Model Interpretation and Debugging**\n    \n    Guo et al. EMNLP 2021\n\n    [arxiv](https://arxiv.org/abs/2012.15781) [github](https://github.com/salesforce/fast-influence-functions)\n\n- **Input Similarity from the Neural Network Perspective**\n    \n    Charpiat et al. NeurIPS 2019\n\n    [arxiv](https://arxiv.org/abs/2102.05262) [github](https://github.com/Lydorn/netsimilarity)\n\n- **RelatIF: Identifying Explanatory Training Examples via Relative Influence**\n\n    Barshan et al. AISTATS 2020\n\n    [arxiv](https://arxiv.org/abs/2003.11630)\n\n- **Representer Point Selection for Explaining Deep Neural Networks**\n- \n    Yeh et al. NeurIPS 2018\n\n    [arxiv](https://arxiv.org/abs/1811.09720)\n\n- **Estimating Training Data Influence by Tracing Gradient Descent**\n  \n    Pruthi et al. NeurIPS 2020\n\n    [arxiv](https://arxiv.org/abs/2002.08484)\n\n- **Scaling Up Influence Functions**\n\n    Schioppa et al. AAAI 2022\n\n    [arxiv](https://arxiv.org/abs/2112.03052) [github](https://github.com/google-research/jax-influence)\n\n- **Interpreting Black Box Predictions using Fisher Kernels**\n\n    Khanna et al. AISTATS 2019\n\n    [arxiv](https://arxiv.org/abs/1810.10118)\n\n\n## [Evaluation](#content)\n- **Evaluation of Similarity-based Explanations**\n\n    Hanawa et al. ICLR 2021\n  \n    [arxiv](https://arxiv.org/abs/2006.04528)\n\n- **An Empirical Comparison of Instance Attribution Methods for NLP**\n    \n    Pezeshkpour et al. NAACL 2021\n\n    [arxiv](https://arxiv.org/abs/2104.04128) [github](https://github.com/successar/instance_attributions_NLP)\n\n- **Explaining Black Box Predictions and Unveiling Data Artifacts through Influence Functions**\n\n    Han et al. ACL 2020\n\n    [arxiv](https://arxiv.org/abs/2005.06676)\n\n- **Influence Functions in Deep Learning are Fragile**\n    \n    Basu et al. ICLR 2021\n\n    [arxiv](https://arxiv.org/abs/2006.14651) \n\n- **Revisiting Methods for Finding Influential Examples**\n    \n    K and Søgaard. AAAI 2022\n    \n    [arxiv](https://arxiv.org/abs/2111.04683)\n\n\n## [Other](#content)\n- **What Neural Networks Memorize and Why: Discovering the Long Tail via Influence Estimation**\n\n    Feldman and Zhang. NeurIPS 2020\n\n    [arxiv](https://arxiv.org/abs/2008.03703)\n\n- **On the Accuracy of Influence Functions for Measuring Group Effects**\n\n    Koh et al.\n\n    [arxiv](https://arxiv.org/abs/1905.13289)\n\n- **Examples are not enough, learn to criticize! Criticism for Interpretability**\n\n    Kim et al. NeurIPS 2016\n\n    [paper](https://papers.nips.cc/paper/2016/hash/5680522b8e2bb01943234bce7bf84534-Abstract.html)\n\n- **Finding Influential Training Samples for Gradient Boosted Decision Trees**\n\n    Sharchilev et al. ICML 2018\n\n    [arxiv](https://arxiv.org/abs/1802.06640)\n\n- **An Empirical Study of Example Forgetting during Deep Neural Network Learnin**\n\n    Toneva et al. ICLR 2019\n\n    [arxiv](https://arxiv.org/abs/1812.05159)\n\n\n- **Towards Efficient Data Valuation Based on the Shapley Value**\n\n    Jia et al. AISTATS 2019\n\n    [arxiv](https://arxiv.org/abs/1902.10275)\n\n- **Data Cleansing for Models Trained with SGD**\n\n    Hara et al. NeurIPS 2019\n\n    [arxiv](https://arxiv.org/abs/1906.08473)\n\n- **Data Shapley: Equitable Valuation of Data for Machine Learning**\n\n    Ghorbani and Zhou. ICML 2019\n\n    [arxiv](https://arxiv.org/abs/1904.02868) [github](https://github.com/amiratag/DataShapley)\n\n- **This Looks Like That: Deep Learning for Interpretable Image Recognition**\n\n    Chen et al. NeurIPS 2019\n\n    [arxiv](https://arxiv.org/abs/1806.10574)\n\n- **Understanding the Origins of Bias in Word Embeddings**\n\n    Brunet et al. ICML 2019\n\n    [arxiv](https://arxiv.org/abs/1810.03611)\n\n- **Data Valuation using Reinforcement Learning**\n\n    Yoon et al. ICML 2020\n\n    [arxiv](https://arxiv.org/abs/1909.11671)\n\n- **Dataset Cartography: Mapping and Diagnosing Datasets with Training Dynamics**\n\n    Swayamdipta et al. EMNLP 2020\n\n    [arxiv](https://arxiv.org/abs/2009.10795)\n\n- **Identifying Mislabeled Data using the Area Under the Margin Ranking**\n\n    Pleiss et al. NeurIPS 2020\n\n    [arxiv](https://arxiv.org/abs/2001.10528)\n\n- **Towards Faithfully Interpretable NLP Systems: How Should We Define and Evaluate Faithfulness?**\n\n    Jacovi and Goldberg. ACL 2020\n\n    [arxiv](https://arxiv.org/abs/2004.03685)\n\n- **Estimating Example Difficulty Using Variance of Gradients**\n\n    Agarwal et al. WHI Workshop @ICML 2020.\n\n    [arxiv](https://arxiv.org/abs/2008.11600)\n\n- **Does learning require memorization? a short tale about a long tail**\n\n    Feldman. STOC 2020\n\n    [arxiv](https://arxiv.org/abs/1906.05271)\n\n- **Multi-Stage Influence Function**\n\n    Chen et al. NeurIPS 2020\n\n    [arxiv](https://arxiv.org/abs/2007.09081)\n\n- **TREX: Tree-Ensemble Representer-Point Explanations**\n\n    Brophy and Lowd. XXAI Workshop at ICML 2020.\n\n    [arxiv](https://arxiv.org/abs/2009.05530)\n\n- **On Second-Order Group Influence Functions for Black-Box Predictions**\n\n    Basu et al. ICML 2020\n\n    [arxiv](https://arxiv.org/abs/1911.00418)\n\n- **On Sample Based Explanation Methods for NLP: Faithfulness, Efficiency and Semantic Evaluation**\n\n    Zhang et al. ACL 2021\n\n    [arxiv](https://arxiv.org/abs/2106.04753)\n\n- **Influence Estimation for Generative Adversarial Networks**\n\n    Terashita et al. ICLR 2021\n\n    [arxiv](https://arxiv.org/abs/2101.08367)\n\n- **Representer Point Selection via Local Jacobian Expansion for Post-hoc Classifier Explanation of Deep Neural Networks and Ensemble Models**\n\n    Sui et al. NeurIPS 2021\n\n    [paper](https://proceedings.neurips.cc//paper/2021/hash/c460dc0f18fc309ac07306a4a55d2fd6-Abstract.html) [github](https://github.com/echoyi/RPS_LJE)\n\n- **Deep Learning on a Data Diet: Finding Important Examples Early in Training**\n\n    Paul et al. NeurIPS 2021\n\n    [arxiv](https://arxiv.org/abs/2107.07075)\n\n- **Understanding Instance-based Interpretability of Variational Auto-Encoders**\n\n    Kong and Chaudhuri. NeurIPS 2021\n\n    [arxiv](https://arxiv.org/abs/2105.14203)\n\n- **Characterizing Structural Regularities of Labeled Data in Overparameterized Models**\n\n    Jiang et al. ICML 2021\n\n    [arxiv](https://arxiv.org/abs/2002.03206)\n\n- **Estimating informativeness of samples with Smooth Unique Information**\n\n    Harutyunyan et al. ICLR 2021\n\n    [arxiv](https://arxiv.org/abs/2101.06640)\n\n- **A Tale Of Two Long Tails**\n\n    D'souza et al. ICML 2021\n\n    [arxiv](https://arxiv.org/abs/2107.13098)\n\n- **HyDRA: Hypergradient Data Relevance Analysis for Interpreting Deep Neural Networks**\n\n    Chen et al. AAAI 2021\n\n    [arxiv](https://arxiv.org/abs/2102.02515)\n\n\n","funding_links":[],"categories":["Other Lists"],"sub_categories":["TeX Lists"],"project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fmaxidl%2Fawesome-instance-attribution","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fmaxidl%2Fawesome-instance-attribution","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fmaxidl%2Fawesome-instance-attribution/lists"}