The DOI matches its Crossref record; 6 of the 6 references it cites were found in a registry or an index.
Make Every Example Count: On the Stability and Utility of Self-Influence for Learning from Noisy NLP Datasets
Open accessAbstract
Increasingly larger datasets have become a standard ingredient to advancing the state-of-theart in NLP.However, data quality might have already become the bottleneck to unlock further gains.Given the diversity and the sizes of modern datasets, standard data filtering is not straight-forward to apply, because of the multifacetedness of the harmful data and elusiveness of filtering rules that would generalize across multiple tasks.We study the fitness of task-agnostic self-influence scores of training examples for data cleaning, analyze their efficacy in capturing naturally occurring outliers, and investigate to what extent self-influence based data cleaning can improve downstream performance in machine translation, question answering and text classification, building up on recent approaches to self-influence calculation and automated curriculum learning.
Overview
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
The check wasn’t confident either way, so it gives no verdict — never disputed.
In the citing paper
“We only select papers published from 2024, minimizing the potential contamination with parametric knowledge during evaluation [25].”
From When AI Co-Scientists Fail: SPOT-a Benchmark for Automated Verification of Scientific Research (2025) · cites this paper as [25]
Experimental — accuracy not yet measuredThe check wasn’t confident either way, so it gives no verdict — never disputed.
In the citing paper
“To minimize contamination against parametric knowledge [25], we aim to include only papers published from 2024 onward.”
From When AI Co-Scientists Fail: SPOT-a Benchmark for Automated Verification of Scientific Research (2025) · cites this paper as [25]
Experimental — accuracy not yet measured
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding
Devlin, J., Chang, M.-W., Lee, K., & Toutanova, K. (2019) · Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers) · doi:10.18653/v1/n19-1423 · Cited by 54,325
Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer
Raffel, C., Shazeer, N., Roberts, A., Lee, K., Narang, S., Matena, M., Zhou, Y., Li, W., & Liu, P. J. (2019) · arXiv · doi:10.48550/arxiv.1910.10683 · Cited by 10,929
Bengio, Y., Louradour, J., Collobert, R., & Weston, J. (2009) · Proceedings of the 26th Annual International Conference on Machine Learning · doi:10.1145/1553374.1553380 · Cited by 5,366
Natural Questions: A benchmark for question answering research
Kwiatkowski, T., Palomaki, J., Redfield, O., Collins, M., Parikh, A., Alberti, C., Epstein, D., Polosukhin, I., Devlin, J., Lee, K., et al. (2019) · Transactions of the Association for Computational Linguistics · doi:10.1162/tacl_a_00276 · Cited by 2,069
TriviaQA: A Large Scale Distantly Supervised Challenge Dataset for Reading Comprehension
Joshi, M., Choi, E., Weld, D., & Zettlemoyer, L. (2017) · Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · doi:10.18653/v1/p17-1147 · Cited by 1,690
Stronger Data Poisoning Attacks Break Data Sanitization Defenses
Koh, P. W., Steinhardt, J., & Liang, P. (2018) · arXiv · doi:10.48550/arxiv.1811.00741 · Cited by 189
Research works printed but not linked: 47
- Peter Auer, Nicolò Cesa-Bianchi, and Paul Fischer. 2002. Finite-time analysis of the multiarmed ban- dit problem. Mach. Learn., 47(2-3):235-256. · DOI · Paper page
- Amittai Axelrod, Xiaodong He, and Jianfeng Gao. 2011. Domain adaptation via pseudo in-domain data selec- tion. In EMNLP.
- Marta Bañón, Pinzhen Chen, Barry Haddow, Kenneth Heafield, Hieu Hoang, Miquel Esplà-Gomis, Mikel L. Forcada, Amir Kamran, Faheem Kirefu, Philipp Koehn, Sergio Ortiz Rojas, Leopoldo Pla Sempere, Gema Ramírez-Sánchez, Elsa Sarrías, Marek Strelec, Brian Thompson, William Waites, Dion Wiggins, and Jaume Zaragoza. 2020. ParaCrawl: Web-scale acquisition of parallel corpora. In ACL. · DOI
- Elnaz Barshan, Marc-Etienne Brunet, and Gintare Karolina Dziugaite. 2020. Relatif: Identifying explanatory training examples via relative influence. In AISTATS. · DOI
- Samyadeep Basu, Philip Pope, and Soheil Feizi. 2021. Influence functions in deep learning are fragile. In ICLR. · DOI
- Yonatan Belinkov and Yonatan Bisk. 2018. Synthetic and natural noise both break neural machine transla- tion. In ICLR. · DOI
- Daniel Borkan, Lucas Dixon, Jeffrey Sorensen, Nithum Thain, and Lucy Vasserman. 2019. Nuanced metrics for measuring unintended bias with real data for text classification. In WWW. · DOI
- Samuel R. Bowman and George Dahl. 2021. What will it take to fix benchmarking in natural language understanding? In NAACL. · DOI
- R. Dennis Cook and Sanford Weisberg. 1980. Char- acterizations of an empirical influence function for detecting influential cases in regression. Technomet- rics, 22(4):495-508.
- Quentin Cormier, Mahdi Milani Fard, Kevin Canini, and Maya Gupta. 2016. Launch and iterate: Reducing prediction churn. In NIPS.
- Sebastian Ebert, Alice Shoshana Jakobovits, and Katja Filippova. 2022. Understanding text classification data and models using aggregated input salience. In R2HCAI workshop at AAAI. · DOI
- Vitaly Feldman and Chiyuan Zhang. 2020. What neural networks memorize and why: Discovering the long tail via influence estimation. In NeurIPS.
- Matt Gardner, Jonathan Berant, Hannaneh Hajishirzi, Alon Talmor, and Sewon Min. 2019. Question an- swering is a format; when is it useful? CoRR, abs/1909.11291. · DOI
- Alex Graves, Marc G. Bellemare, Jacob Menick, Rémi Munos, and Koray Kavukcuoglu. 2017. Automated curriculum learning for neural networks. In ICML.
- Han Guo, Nazneen Fatema Rajani, Peter Hase, Mohit Bansal, and Caiming Xiong. 2021. Fastif: Scalable influence functions for efficient model interpretation and debugging. In EMNLP. · DOI
- Mandy Guo, Joshua Ainslie, David Uthus, Santiago On- tanon, Jianmo Ni, Yun-Hsuan Sung, and Yinfei Yang. 2022. LongT5: Efficient text-to-text transformer for long sequences. In NAACL. · DOI · Paper page
- Xiaochuang Han, Byron C. Wallace, and Yulia Tsvetkov. 2020. Explaining black box predictions and unveil- ing data artifacts through influence functions. In ACL. · DOI
- Lu Jiang, Di Huang, Mason Liu, and Weilong Yang. 2020. Beyond synthetic noise: Deep learning on controlled noisy labels. In ICML.
- Karthikeyan K and Anders Søgaard. 2021. Revisit- ing methods for finding influential examples. CoRR, abs/2111.04683. · DOI
- Huda Khayrallah and Philipp Koehn. 2018. On the impact of various types of noise on neural machine translation. In WMT. · DOI
- Pang Wei Koh and Percy Liang. 2017. Understanding black-box predictions via influence functions. In ICML. · DOI · Paper page
- Shuming Kong, Yanyan Shen, and Linpeng Huang. 2022. Resolving training biases via influence-based data relabeling. In ICLR.
- Julia Kreutzer, David Vilar, and Artem Sokolov. 2021. Bandits don't follow rules: Balancing multi-facet machine translation with multi-armed bandits. In EMNLP.
- Ankit Kumar, Piyush Makhija, and Anuj Gupta. 2020. Noisy text data: Achilles' heel of BERT. In Work- shop on Noisy User-generated Text. · DOI
- Gaurav Kumar, George Foster, Colin Cherry, and Maxim Krikun. 2019. Reinforcement learning based curriculum optimization for neural machine transla- tion. In NAACL. · DOI
- Tsz Kin Lam, Eva Hasler, and Felix Hieber. 2022. An- alyzing the use of influence functions for instance- specific data filtering in neural machine translation. In WMT. · DOI
- Duc Tam Nguyen, Chaithanya Kumar Mummadi, Thi Phuong Nhung Ngo, Thi Hoai Phuong Nguyen, Laura Beggel, and Thomas Brox. 2020. Self: Learn- ing to filter noisy labels with self-ensembling. In ICLR.
- Barak A. Pearlmutter. 1994. Fast Exact Multiplication by the Hessian. Neural Computation, 6(1):147-160.
- Denis Peskov, Joe Barrow, Pedro Rodriguez, Graham Neubig, and Jordan Boyd-Graber. 2019. Mitigating noisy inputs for question answering. In Interspeech. · DOI
- Emmanouil Antonios Platanios, Otilia Stretcu, Graham Neubig, Barnabas Poczos, and Tom M. Mitchell. 2019. Competence-based curriculum learning for neural machine translation. In NAACL. · DOI
- Geoff Pleiss, Tianyi Zhang, Ethan R. Elenberg, and Kilian Q. Weinberger. 2020. Identifying mislabeled data using the area under the margin ranking. In NeurIPS. · DOI
- Garima Pruthi, Frederick Liu, Mukund Sundararajan, and Satyen Kale. 2020. Estimating training data influence by tracing gradient descent. In NeurIPS. · DOI
- Adam Roberts, Hyung Won Chung, Anselm Levskaya, Gaurav Mishra, James Bradbury, Daniel Andor, Sha- ran Narang, Brian Lester, Colin Gaffney, Afroz Mohiuddin, Curtis Hawthorne, Aitor Lewkowycz, Alex Salcianu, Marc van Zee, Jacob Austin, Sebas- tian Goodman, Livio Baldini Soares, Haitang Hu, Sasha Tsvyashchenko, Aakanksha Chowdhery, Jas- mijn Bastings, Jannis Bulian, Xavier Garcia, Jianmo Ni, Andrew Chen, Kathleen Kenealy, Jonathan H. Clark, Stephan Lee, Dan Garrette, James Lee-Thorp, Colin Raffel, Noam Shazeer, Marvin Ritter, Maarten Bosma, Alexandre Passos, Jeremy Maitin-Shepard, Noah Fiedel, Mark Omernick, Brennan Saeta, Ryan Sepassi, Alexander Spiridonov, Joshua Newlan, and Andrea Gesmundo. 2022. Scaling up models and data with t5x and seqio. CoRR, abs/2203.17189. · DOI
- Anna Rogers, Matt Gardner, and Isabelle Augenstein. 2023. QA dataset explosion: A taxonomy of NLP resources for question answering and reading com- prehension. ACM Computing Surveys, 55(10):1-45. · DOI
- David Rolnick, Andreas Veit, Serge Belongie, and Nir Shavit. 2017. Deep learning is robust to massive label noise. CoRR, abs/1705.10694. · DOI · Paper page
- Andrea Schioppa, Katja Filippova, Ivan Titov, and Polina Zablotskaia. 2023. Theoretical and practi- cal perspectives on what influence functions do. In NeurIPS.
- Andrea Schioppa, Polina Zablotskaia, David Vilar, and Artem Sokolov. 2021. Scaling up influence functions. In AAAI.
- Saku Sugawara, Kentaro Inui, Satoshi Sekine, and Akiko Aizawa. 2018. What makes reading compre- hension questions easier? In EMNLP. · DOI
- Swabha Swayamdipta, Roy Schwartz, Nicholas Lourie, Yizhong Wang, Hannaneh Hajishirzi, Noah A. Smith, and Yejin Choi. 2020. Dataset cartography: Mapping and diagnosing datasets with training dynamics. In EMNLP. · DOI
- Mariya Toneva, Alessandro Sordoni, Remi Tachet des Combes, Adam Trischler, Yoshua Bengio, and Geof- frey J. Gordon. 2019. An empirical study of example forgetting during deep neural network learning. In ICLR. · DOI
- Vaibhav Vaibhav, Sumeet Singh, Craig Stewart, and Gra- ham Neubig. 2019. Improving robustness of machine translation with synthetic noise. In NAACL. · DOI
- Marlies van der Wees, Arianna Bisazza, and Christof Monz. 2017. Dynamic data selection for neural ma- chine translation. In EMNLP. · DOI
- Ashish Vaswani, Noam Shazeer, Niki Parmar, Jakob Uszkoreit, Llion Jones, Aidan N. Gomez, Lukasz Kaiser, and Illia Polosukhin. 2017. Attention is all you need. In NIPS. · DOI · Paper page
- Wei Wang, Taro Watanabe, Macduff Hughes, Tetsuji Nakagawa, and Ciprian Chelba. 2018. Denoising neural machine translation training with trusted data and online data selection. In WMT. · DOI
- Ellery Wulczyn, Nithum Thain, and Lucas Dixon. 2017. Ex machina: Personal attacks seen at scale. In WWW. · DOI
- Chih-Kuan Yeh, Ankur Taly, Mukund Sundararajan, Frederick Liu, and Pradeep Ravikumar. 2022. First is better than last for language data influence. In NeurIPS. · DOI
- Chiyuan Zhang, Samy Bengio, Moritz Hardt, Benjamin Recht, and Oriol Vinyals. 2017. Understanding deep learning requires rethinking generalization. In ICLR. · DOI · Paper page
Cited by
Sorted by citations
When AI Co-Scientists Fail: SPOT-a Benchmark for Automated Verification of Scientific Research
Son, G., Hong, J., Fan, H., Nam, H., Ko, H., Lim, S., Song, J., Choi, J., Paulo, G., Yu, Y., et al. (2025) · arXiv · doi:10.48550/arxiv.2505.11855 · Cited by 5
Metadata
- Authors
- Bejan, I., Sokolov, A., & Filippova, K.
- ORCID
- Filippova, Katja
- Venue
- Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing
- Published
- 2023
- Pages
- 10107-10121
- OpenAlex
- W4389519087
- Cited by
- 7
- References
- 6
- Citations by year
- 2026: 1 · 2025: 2 · 2024: 1 · 2023: 3
- FWCI
- 1.24
- Type
- conference paper
- Language
- en
- Open access
- hybrid
- doi.org
- Licence
- cc-by
- Locations
- Conference on Empirical Methods in Natural Language Processing (EMNLP) (publishedVersion, cc-by)
- arXiv (Cornell University) (submittedVersion, cc-by)
- arXiv (Cornell University) (cc-by)
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › Machine Learning and Data Classification
- Physical Sciences › Computer Science › Artificial Intelligence › Natural Language Processing Techniques
- Physical Sciences › Computer Science › Artificial Intelligence › Topic Modeling
- Keywords
- self-influence, noisy training data, data cleaning, data quality, outlier detection, machine translation, question answering, text classification
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
