English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Tools annual

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

The prior work on natural language inference (NLI) debiasing mainly targetsat 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 explorehow to (or is it possible to) make the NLI models robust to multiple distinctadversarial attacks while keeping or even strengthening the models'generalization power. We firstly benchmark prevailing neural NLI modelsincluding pretrained ones on various adversarial datasets. We then try tocombat distinct known biases by modifying a mixture of experts (MoE) ensemblemethod and show that it's nontrivial to mitigate multiple NLI biases at thesame time, and that model-level ensemble method outperforms MoE ensemblemethod. We also perform data augmentation including text swap, wordsubstitution and paraphrase and prove its efficiency in combating various(though not all) adversarial attacks at the same time. Finally, we investigateseveral methods to merge heterogeneous training data (1.35M) and perform modelensembling, which are straightforward but effective to strengthen NLI models.

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