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

Intent Classification (IC) and Slot Labeling (SL) models, which form thebasis of dialogue systems, often encounter noisy data in real-wordenvironments. In this work, we investigate how robust IC/SL models are to noisydata. We collect and publicly release a test-suite for seven common noise typesfound in production human-to-bot conversations (abbreviations, casing,misspellings, morphological variants, paraphrases, punctuation and synonyms).On this test-suite, we show that common noise types substantially degrade theIC accuracy and SL F1 performance of state-of-the-art BERT-based IC/SL models.By leveraging cross-noise robustness transfer -- training on one noise type toimprove robustness on another noise type -- we design aggregatedata-augmentation approaches that increase the model performance across allseven noise types by +10.8% for IC accuracy and +15 points for SL F1 onaverage. To the best of our knowledge, this is the first work to present asingle IC/SL model that is robust to a wide range of noise phenomena.

On the Robustness of Intent Classification and Slot Labeling in Goal-oriented Dialog Systems to Real-world Noise