StereoTales shows that all tested LLMs emit harmful stereotypes in open-ended stories, with associations adapting to prompt language and targeting locally salient groups rather than transferring uniformly across languages.
arXiv preprint arXiv:2306.11507 (2023)
4 Pith papers cite this work. Polarity classification is still indexing.
4
Pith papers citing it
citation-role summary
background 1
method 1
citation-polarity summary
years
2026 4representative citing papers
LLM-based conversational interface for Android reduces task time and mental effort for blind users versus traditional gesture-based screen readers like TalkBack.
MiJaBench shows LLM safety refusals vary by up to 33 percentage points within one model depending only on the targeted demographic.
A multi-role red-teaming framework with attacker, target, and jury LLMs measures faithfulness in English and Arabic, finding false-premise prompts and length limits change unfaithfulness rates.
citing papers explorer
-
StereoTales: A Multilingual Framework for Open-Ended Stereotype Discovery in LLMs
StereoTales shows that all tested LLMs emit harmful stereotypes in open-ended stories, with associations adapting to prompt language and targeting locally salient groups rather than transferring uniformly across languages.
-
Insight: Enhancing Mobile Accessibility for Blind and Visually Impaired Users with LLMs
LLM-based conversational interface for Android reduces task time and mental effort for blind users versus traditional gesture-based screen readers like TalkBack.
-
Safety Is Not Universal: The Selective Safety Trap in LLM Alignment
MiJaBench shows LLM safety refusals vary by up to 33 percentage points within one model depending only on the targeted demographic.
-
A Red Teaming Framework for Large Language Models: A Case Study on Faithfulness Evaluation
A multi-role red-teaming framework with attacker, target, and jury LLMs measures faithfulness in English and Arabic, finding false-premise prompts and length limits change unfaithfulness rates.