Papers with Code · CC BY-SA 4.0

Bias Detection

19 평가 기록 5 datasets

Bias detection is the task of detecting and measuring racism, sexism and otherwise discriminatory behavior in a model (Source: https://stereoset.mit.edu/)

논문

StereoSet: Measuring stereotypical bias in pretrained language models

2004.09456 2020-04-20 이 페이지의 평가 8건

ICAT Score: 72.97

Benchmarking Llama2, Mistral, Gemma and GPT for Factuality, Toxicity, Bias and Propensity for Hallucinations

2404.09785 2024-04-15 이 페이지의 평가 5건

Best-of: 0.5

Galactica: A Large Language Model for Science

2211.09085 2022-11-16 이 페이지의 평가 3건

ICAT Score: 65.6LMS: 75SS: 56.2

BAD: BiAs Detection for Large Language Models in the context of candidate screening

2305.10407 2023-05-17 이 페이지의 평가 1건

ICAT Score: 23.44

Towards Detection of Subjective Bias using Contextualized Word Embeddings

2002.06644 2020-02-16 이 페이지의 평가 1건

F1: 70.4

Uncovering bias in the PlantVillage dataset

2206.04374 2022-06-09 이 페이지의 평가 1건

Accuracy (%): 49.0