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 Tools annual

Global: Zendy Free Plan

Global: Zendy Tools monthly

Global: Zendy Plus monthly

As Large Language Models (LLMs) have risen in prominence over the past fewyears, there has been concern over the potential biases in LLMs inherited fromthe training data. Previous studies have examined how LLMs exhibit implicitbias, such as when response generation changes when different social contextsare introduced. We argue that this implicit bias is not only an ethical, butalso a technical issue, as it reveals an inability of LLMs to accommodateextraneous information. However, unlike other measures of LLM intelligence,there are no standard methods to benchmark this specific subset of LLM bias. Tobridge this gap, we developed a method for calculating an easily interpretablebenchmark, DIF (Demographic Implicit Fairness), by evaluating preexisting LLMlogic and math problem datasets with sociodemographic personas. We demonstratethat this method can statistically validate the presence of implicit bias inLLM behavior and find an inverse trend between question answering accuracy andimplicit bias, supporting our argument.

DIF: A Framework for Benchmarking and Verifying Implicit Bias in LLMs