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

For many real-world applications, the user-generated inputs usually containvarious noises due to speech recognition errors caused by linguisticvariations1 or typographical errors (typos). Thus, it is crucial to test modelperformance on data with realistic input noises to ensure robustness andfairness. However, little study has been done to construct such benchmarks forChinese, where various language-specific input noises happen in the real world.In order to fill this important gap, we construct READIN: a Chinese multi-taskbenchmark with REalistic And Diverse Input Noises. READIN contains four diversetasks and requests annotators to re-enter the original test data with twocommonly used Chinese input methods: Pinyin input and speech input. We designedour annotation pipeline to maximize diversity, for example by instructing theannotators to use diverse input method editors (IMEs) for keyboard noises andrecruiting speakers from diverse dialectical groups for speech noises. Weexperiment with a series of strong pretrained language models as well as robusttraining methods, we find that these models often suffer significantperformance drops on READIN even with robustness methods like dataaugmentation. As the first large-scale attempt in creating a benchmark withnoises geared towards user-generated inputs, we believe that READIN serves asan important complement to existing Chinese NLP benchmarks. The source code anddataset can be obtained from https://github.com/thunlp/READIN.

READIN: A Chinese Multi-Task Benchmark with Realistic and Diverse Input Noises