bookCOFFEE: Counterfactual Fairness for Personalized Text Generation in Explainable Recommendation
… that sentiment analysis systems yield different sentiment scores for sentences containi...
2022-10-14 00:00:00
WangY.C.SanjabiM.NieS.WangN.WangQ.
bookEr-test: Evaluating explanation regularization methods for nlp models
… Functional Tests Figure 4 demonstrates the failure rates on functional tests (as listed...
2022-05-25 00:00:00
LiuZ.SanjabiM.NieS.JoshiB.ChanA.
bookEr-test: Evaluating explanation regularization methods for language models
… We observe (Table 2) the drop in performance (∆) for sentiment analysis and NLI when ...
2022-01-01 00:00:00
bookDetection, disambiguation, re-ranking: autoregressive entity linking as a multi-task problem
We propose an autoregressive entity linking model, that is trained with two auxiliary tasks, and lea...
GuJ.SanjabiM.NieS.MriniK.WangS.
bookMSD: Saliency-aware Knowledge Distillation for Multimodal Understanding
To reduce a model size but retain performance, we often rely on knowledge distillation (KD) which tr...
2021-01-01 00:00:00
JinW.SanjabiM.NieS.TanL.RenX.
bookModality-specific distillation
… The task is a binary classification problem, which is to detect hate speech in multimod...