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

Recently, significant public efforts have been directed towards developinglow-cost models with capabilities akin to ChatGPT, thereby fostering the growthof open-source conversational models. However, there remains a scarcity ofcomprehensive and in-depth evaluations of these models' performance. In thisstudy, we examine the influence of training data factors, including quantity,quality, and linguistic distribution, on model performance. Our analysis isgrounded in several publicly accessible, high-quality instruction datasets, aswell as our own Chinese multi-turn conversations. We assess various modelsusing a evaluation set of 1,000 samples, encompassing nine real-worldscenarios. Our goal is to supplement manual evaluations with quantitativeanalyses, offering valuable insights for the continued advancement ofopen-source chat models. Furthermore, to enhance the performance and trainingand inference efficiency of models in the Chinese domain, we extend thevocabulary of LLaMA - the model with the closest open-source performance toproprietary language models like GPT-3 - and conduct secondary pre-training on3.4B Chinese words. We make our model, data, as well as code publiclyavailable.

Towards Better Instruction Following Language Models for Chinese: Investigating the Impact of Training Data and Evaluation