An Empirical Study on Model-agnostic Debiasing Strategies for Robust Natural Language Inference

10/08/2020
by   Tianyu Liu, et al.
6

The prior work on natural language inference (NLI) debiasing mainly targets at one or few known biases while not necessarily making the models more robust. In this paper, we focus on the model-agnostic debiasing strategies and explore how to (or is it possible to) make the NLI models robust to multiple distinct adversarial attacks while keeping or even strengthening the models' generalization power. We firstly benchmark prevailing neural NLI models including pretrained ones on various adversarial datasets. We then try to combat distinct known biases by modifying a mixture of experts (MoE) ensemble method and show that it's nontrivial to mitigate multiple NLI biases at the same time, and that model-level ensemble method outperforms MoE ensemble method. We also perform data augmentation including text swap, word substitution and paraphrase and prove its efficiency in combating various (though not all) adversarial attacks at the same time. Finally, we investigate several methods to merge heterogeneous training data (1.35M) and perform model ensembling, which are straightforward but effective to strengthen NLI models.

READ FULL TEXT

page 3

page 8

research
02/28/2023

SMoA: Sparse Mixture of Adapters to Mitigate Multiple Dataset Biases

Recent studies reveal that various biases exist in different NLP tasks, ...
research
06/20/2020

Defense against Adversarial Attacks in NLP via Dirichlet Neighborhood Ensemble

Despite neural networks have achieved prominent performance on many natu...
research
06/08/2022

Adversarial Text Normalization

Text-based adversarial attacks are becoming more commonplace and accessi...
research
04/16/2020

There is Strength in Numbers: Avoiding the Hypothesis-Only Bias in Natural Language Inference via Ensemble Adversarial Training

Natural Language Inference (NLI) datasets contain annotation artefacts r...
research
09/13/2019

simple but effective techniques to reduce biases

There have been several studies recently showing that strong natural lan...
research
11/03/2020

Detecting Word Sense Disambiguation Biases in Machine Translation for Model-Agnostic Adversarial Attacks

Word sense disambiguation is a well-known source of translation errors i...
research
06/16/2020

Model Agnostic Combination for Ensemble Learning

Ensemble of models is well known to improve single model performance. We...

Please sign up or login with your details

Forgot password? Click here to reset