This is the official implementation for the experiments in our paper Convex Surrogates for Unbiased Loss Functions in Extreme Classification With Missing Labels
To run the experiments using propensity-weighted variants of APLC_XLNet [1], AttentionXML [2], and DiSMEC [3], please read the instructions in the corresponding folders.
[1] Ye et al., Pretrained Generalized Autoregressive Model with Adaptive Probabilistic Label Clusters for Extreme Multi-label Text Classification, ICML 2020.
[2] You et al., AttentionXML: Label Tree-based Attention-Aware Deep Model for High-Performance Extreme Multi-Label Text Classification, NeurIPS 2019.
[3] R. Babbar, B. Schölkopf, DiSMEC: Distributed Sparse Machines for Extreme Multi-label Classification, WSDM 2017.
[4] M. Qaraei et al., Convex Surrogates for Unbiased Loss Functions in Extreme Classification With Missing Labels, WWW (2021).