SCOPE is a new large-scale dataset of counterfactual prompt pairs for evaluating fairness and stereotype sensitivity in LLMs across 1,438 topics, nine bias dimensions, 1,536 groups, and four communicative intents.
Stereoset: Measuring stereotyp- ical bias in pretrained language models,
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
fields
cs.SE 2years
2026 2verdicts
UNVERDICTED 2representative citing papers
Sensitive prompts serve as an early-warning signal for fairness risks in LLMs by eliciting responses that often miss ethical or contextual implications.
citing papers explorer
-
SCOPE: A Dataset of Stereotyped Prompts for Counterfactual Fairness Assessment of LLMs
SCOPE is a new large-scale dataset of counterfactual prompt pairs for evaluating fairness and stereotype sensitivity in LLMs across 1,438 topics, nine bias dimensions, 1,536 groups, and four communicative intents.
-
Bias Ahead: Sensitive Prompts as Early Warnings for Fairness in Large Language Models
Sensitive prompts serve as an early-warning signal for fairness risks in LLMs by eliciting responses that often miss ethical or contextual implications.