Papers › Text Classification in the Wild: a Large-scale Long-tailed Name Normalization Dataset

Text Classification in the Wild: a Large-scale Long-tailed Name Normalization Dataset

19 Feb 2023arXiv:2302.09509archive 2025-07-28

Jiexing Qi, Shuhao Li, Zhixin Guo, Yusheng Huang, Chenghu Zhou, Weinan Zhang, Xinbing Wang, Zhouhan Lin

Real-world data usually exhibits a long-tailed distribution,with a few frequent labels and a lot of few-shot labels. The study of institution name normalization is a perfect application case showing this phenomenon. There are many institutions worldwide with enormous variations of their names in the publicly available literature. In this work, we first collect a large-scale institution name normalization dataset LoT-insts1, which contains over 25k classes that exhibit a naturally long-tailed distribution. In order to isolate the few-shot and zero-shot learning scenarios from the massive many-shot classes, we construct our test set from four different subsets: many-, medium-, and few-shot sets, as well as a zero-shot open set. We also replicate several important baseline methods on our data, covering a wide range from search-based methods to neural network methods that use the pretrained BERT model. Further, we propose our specially pretrained, BERT-based model that shows better out-of-distribution generalization on few-shot and zero-shot test sets. Compared to other datasets focusing on the long-tailed phenomenon, our dataset has one order of magnitude more training data than the largest existing long-tailed datasets and is naturally long-tailed rather than manually synthesized. We believe it provides an important and different scenario to study this problem. To our best knowledge, this is the first natural language dataset that focuses on long-tailed and open-set classification problems.

PaperPDFCode

Code

lumia-group/lot-insts officialmentioned in paperpytorch report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

Long-tail LearningOut-of-Distribution GeneralizationText ClassificationZero-Shot Learningopen-set classificationtext-classification

Datasets

Introduced by this paper, per the archive.

Lot-insts

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Long-tail Learning Lot-insts Character-BERT+RS Macro-F1 65.90 #1 of 1 Archive leaderboard report
Text Classification Lot-insts Character-BERT+RS Accuracy 83.73 #1 of 5 Archive leaderboard report
Text Classification Lot-insts Character-BERT+RS Macro-F1 65.9 #1 of 5 Archive leaderboard report
Text Classification Lot-insts CD-V1 Accuracy 79.97 #2 of 5 Archive leaderboard report
Text Classification Lot-insts CD-V1 Macro-F1 59.64 #2 of 5 Archive leaderboard report
Text Classification Lot-insts sCool Accuracy 76.72 #3 of 5 Archive leaderboard report
Text Classification Lot-insts sCool Macro-F1 52.41 #3 of 5 Archive leaderboard report
Text Classification Lot-insts FastText Accuracy 74.93 #4 of 5 Archive leaderboard report
Text Classification Lot-insts FastText Macro-F1 44.38 #4 of 5 Archive leaderboard report
Text Classification Lot-insts Naive Bayes Accuracy 72.2 #5 of 5 Archive leaderboard report
Text Classification Lot-insts Naive Bayes Macro-F1 50.2 #5 of 5 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Methods

AdamAttentionAttention DropoutBERTDense ConnectionsDropoutLayer NormalizationLinear LayerLinear Warmup With Linear DecayMulti-Head AttentionResidual ConnectionSoftmaxTestWeight DecayWordPiece

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections