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/

Zendy Plus

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

Zendy Tools

Zendy Open

Recent literature has suggested the potential of using large language models(LLMs) to make classifications for tabular tasks. However, LLMs have been shownto exhibit harmful social biases that reflect the stereotypes and inequalitiespresent in society. To this end, as well as the widespread use of tabular datain many high-stake applications, it is important to explore the followingquestions: what sources of information do LLMs draw upon when makingclassifications for tabular tasks; whether and to what extent are LLMclassifications for tabular data influenced by social biases and stereotypes;and what are the consequential implications for fairness?  Through a series of experiments, we delve into these questions and show thatLLMs tend to inherit social biases from their training data which significantlyimpact their fairness in tabular classification tasks. Furthermore, ourinvestigations show that in the context of bias mitigation, though in-contextlearning and finetuning have a moderate effect, the fairness metric gap betweendifferent subgroups is still larger than that in traditional machine learningmodels, such as Random Forest and shallow Neural Networks. This observationemphasizes that the social biases are inherent within the LLMs themselves andinherited from their pretraining corpus, not only from the downstream taskdatasets. Besides, we demonstrate that label-flipping of in-context examplescan significantly reduce biases, further highlighting the presence of inherentbias within LLMs.

Confronting LLMs with Traditional ML: Rethinking the Fairness of Large Language Models in Tabular Classifications