Keyword

feature ablation

Description: An interpretability technique that removes or suppresses input features to measure their contribution to a model’s predictions.
OpenAlex ID: keywords/feature-ablation
2024 · Marks, Samuel, Rager, Can, et al. · arXiv (Cornell University) · 365
2023 · Yicheng Cai, Haizhou Wang, et al. · Expert Systems with Applications · 88
2024 · Jinwei Hu, Kewei Zhu, et al. · Chemical Engineering Journal · 59
2025 · Benedikt Heudorfer, Hoshin V. Gupta, et al. · Geophysical Research Letters · 48
2025 · Boyi Deng, Yu Wan, et al. · Proceedings of the conference - Association for Computational Linguistics. Meeting · 44
2022 · Xingchen Wang, Tianqi Zhu, et al. · Frontiers in Cardiovascular Medicine · 40
2017 · Mowery, Danielle, Bryan, Craig, et al. · arXiv (Cornell University) · 40
1996 · Joseph F. McCarthy, Wendy G. Lehnert · ScholarWorks@UMassAmherst (University of Massachusetts Amherst) · 32
2024 · Yuxin Liu, Bin Yang, et al. · Journal of Hazardous Materials · 31
2019 · Merrick, Luke · arXiv (Cornell University) · 28
2021 · Justus Mattern, Yu Qiao, et al. · Fact Extraction and VERification Workshop (FEVER) · 25
2019 · Mingyu Derek Ma, Kevin Bowden, et al. · Proceedings of the conference - Association for Computational Linguistics. Meeting · 23
2023 · Foley, Myles, Wang, Mia, et al. · arXiv (Cornell University) · 22
2025 · Xiaoqing Sun, Alessandro Stolfo, et al. · neural information processing systems · 19
2024 · Karvonen, Adam, Rager, Can, et al. · arXiv (Cornell University) · 19
2024 · Nawshad Farruque, Randy Goebel, et al. · Natural Language Processing Journal · 19
2024 · Davide Italo Serramazza, Thach Le Nguyen, et al. · Lecture notes in computer science · 14
2020 · Georgios Barlas, Christos Veinidis, et al. · The Visual Computer · 14
2022 · Sourabh Zanwar, Daniel Wiechmann, et al. · International Workshop on Health Text Mining and Information Analysis (LOUHI) · 13
2025 · Leonid Komissarov, Nenad Manevski, et al. · Journal of Chemical Information and Modeling · 11
Works count: 344
Citation count: 1,331
Year
Topic
Type