Papers › LOGAN: Local Group Bias Detection by Clustering

LOGAN: Local Group Bias Detection by Clustering

6 Oct 2020EMNLP 2020 11arXiv:2010.02867archive 2025-07-28

Jieyu Zhao, Kai-Wei Chang

Machine learning techniques have been widely used in natural language processing (NLP). However, as revealed by many recent studies, machine learning models often inherit and amplify the societal biases in data. Various metrics have been proposed to quantify biases in model predictions. In particular, several of them evaluate disparity in model performance between protected groups and advantaged groups in the test corpus. However, we argue that evaluating bias at the corpus level is not enough for understanding how biases are embedded in a model. In fact, a model with similar aggregated performance between different groups on the entire data may behave differently on instances in a local region. To analyze and detect such local bias, we propose LOGAN, a new bias detection technique based on clustering. Experiments on toxicity classification and object classification tasks show that LOGAN identifies bias in a local region and allows us to better analyze the biases in model predictions.

PaperPDFConference PDFCode

In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

Code

uclanlp/clusters officialmentioned in papermentioned on GitHubpytorch 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

BIG-bench Machine LearningBias DetectionClustering

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

Methods

1x1 ConvolutionAdamBatch NormalizationBigGAN-deepBottleneck Residual BlockConditional Batch NormalizationConvolutionDense ConnectionsEarly StoppingEuclidean Norm RegularizationFeedforward NetworkGAN Hinge LossLOGANLatent OptimisationLinear LayerNatural Gradient DescentNon-Local BlockNon-Local OperationOff-Diagonal Orthogonal RegularizationProjection DiscriminatorReLUResidual ConnectionSAGANSoftmaxSpectral NormalizationTTURTruncation Trick

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