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 Plus annual

Presents the access of premium content as premium feature

Premium Content

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 Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Behavior of deep neural networks can be inconsistent between differentversions. Regressions during model update are a common cause of concern thatoften over-weigh the benefits in accuracy or efficiency gain. This work focuseson quantifying, reducing and analyzing regression errors in the NLP modelupdates. Using negative flip rate as regression measure, we show thatregression has a prevalent presence across tasks in the GLUE benchmark. Weformulate the regression-free model updates into a constrained optimizationproblem, and further reduce it into a relaxed form which can be approximatelyoptimized through knowledge distillation training method. We empiricallyanalyze how model ensemble reduces regression. Finally, we conduct CheckListbehavioral testing to understand the distribution of regressions acrosslinguistic phenomena, and the efficacy of ensemble and distillation methods.

Regression Bugs Are In Your Model! Measuring, Reducing and Analyzing Regressions In NLP Model Updates