paper-with-me

홈 › Papers

We Need to Talk About Random Splits

2020-05-01 · EACL 2021 2 · Anders Søgaard, Sebastian Ebert, Jasmijn Bastings, Katja Filippova

Gorman and Bedrick (2019) argued for using random splits rather than standard splits in NLP experiments. We argue that random splits, like standard splits, lead to overly optimistic performance estimates. We can also split data in biased or adversarial ways, e.g., training on short sentences and evaluating on long ones. Biased sampling has been used in domain adaptation to simulate real-world drift; this is known as the covariate shift assumption. In NLP, however, even worst-case splits, maximizing bias, often under-estimate the error observed on new samples of in-domain data, i.e., the data that models should minimally generalize to at test time. This invalidates the covariate shift assumption. Instead of using multiple random splits, future benchmarks should ideally include multiple, independent test sets instead; if infeasible, we argue that multiple biased splits leads to more realistic performance estimates than multiple random splits.

📄 PDF Abstract BibTeX arXiv:2005.00636

Code (1)

google-research/google-research/tree/master/talk_about_random_splits 공식 구현 tf

Tasks

Domain Adaptation

Similar Papers 제목 키워드 기반

We Need to Talk about Standard Splits

2019-07-01 · ACL 2019 7 · Kyle Gorman, Steven Bedrick

It is standard practice in speech {\&} language technology to rank systems according to their performance on a test set held out for evaluation. However, few researchers apply statistical tests to determine whether diffe…

We Need to Talk About train-dev-test Splits

2021-11-01 · EMNLP 2021 11 · Rob van der Goot

Standard train-dev-test splits used to benchmark multiple models against each other are ubiquitously used in Natural Language Processing (NLP). In this setup, the train data is used for training the model, the developmen…

Model Selection

It's about Time: Rethinking Evaluation on Rumor Detection Benchmarks using Chronological Splits

2023-02-06 · Yida Mu, Kalina Bontcheva, Nikolaos Aletras

New events emerge over time influencing the topics of rumors in social media. Current rumor detection benchmarks use random splits as training, development and test sets which typically results in topical overlaps. Conse…

We need to talk about random seeds

2022-10-24 · Steven Bethard

Modern neural network libraries all take as a hyperparameter a random seed, typically used to determine the initial state of the model parameters. This opinion piece argues that there are some safe uses for random seeds:…

We need to talk about random seeds

2021-11-16 · ACL ARR September 2021 9 · Anonymous

Modern neural network libraries all take as a hyperparameter a random seed, typically used to determine the initial state of the model parameters. This position piece argues that there are some safe uses for random seeds…

Position