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

With the rapid evolution of large language models (LLMs), there is a growingconcern that they may pose risks or have negative social impacts. Therefore,evaluation of human values alignment is becoming increasingly important.Previous work mainly focuses on assessing the performance of LLMs on certainknowledge and reasoning abilities, while neglecting the alignment to humanvalues, especially in a Chinese context. In this paper, we present CValues, thefirst Chinese human values evaluation benchmark to measure the alignmentability of LLMs in terms of both safety and responsibility criteria. As aresult, we have manually collected adversarial safety prompts across 10scenarios and induced responsibility prompts from 8 domains by professionalexperts. To provide a comprehensive values evaluation of Chinese LLMs, we notonly conduct human evaluation for reliable comparison, but also constructmulti-choice prompts for automatic evaluation. Our findings suggest that whilemost Chinese LLMs perform well in terms of safety, there is considerable roomfor improvement in terms of responsibility. Moreover, both the automatic andhuman evaluation are important for assessing the human values alignment indifferent aspects. The benchmark and code is available on ModelScope andGithub.

CValues: Measuring the Values of Chinese Large Language Models from Safety to Responsibility