Irrelevant Alternatives Bias Large Language Model Hiring Decisions
Fuente:
arXiv
Salvato in:
| Autori principali: | , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2024
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866917850766114816 |
|---|---|
| author | Valkanova, Kremena Yordanov, Pencho |
| author_facet | Valkanova, Kremena Yordanov, Pencho |
| contents | We investigate whether LLMs display a well-known human cognitive bias, the attraction effect, in hiring decisions. The attraction effect occurs when the presence of an inferior candidate makes a superior candidate more appealing, increasing the likelihood of the superior candidate being chosen over a non-dominated competitor. Our study finds consistent and significant evidence of the attraction effect in GPT-3.5 and GPT-4 when they assume the role of a recruiter. Irrelevant attributes of the decoy, such as its gender, further amplify the observed bias. GPT-4 exhibits greater bias variation than GPT-3.5. Our findings remain robust even when warnings against the decoy effect are included and the recruiter role definition is varied. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2409_15299 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Irrelevant Alternatives Bias Large Language Model Hiring Decisions Valkanova, Kremena Yordanov, Pencho Computers and Society Artificial Intelligence Human-Computer Interaction We investigate whether LLMs display a well-known human cognitive bias, the attraction effect, in hiring decisions. The attraction effect occurs when the presence of an inferior candidate makes a superior candidate more appealing, increasing the likelihood of the superior candidate being chosen over a non-dominated competitor. Our study finds consistent and significant evidence of the attraction effect in GPT-3.5 and GPT-4 when they assume the role of a recruiter. Irrelevant attributes of the decoy, such as its gender, further amplify the observed bias. GPT-4 exhibits greater bias variation than GPT-3.5. Our findings remain robust even when warnings against the decoy effect are included and the recruiter role definition is varied. |
| title | Irrelevant Alternatives Bias Large Language Model Hiring Decisions |
| topic | Computers and Society Artificial Intelligence Human-Computer Interaction |
| url | https://arxiv.org/abs/2409.15299 |