• Evaluated 11 LLMs generating six-word subject lines for German counselling emails. • Used hierarchical assessment: first categorize outputs, then rank within categories. • Nine a
• Computer Science > Machine Learning [Submitted on 28 Jan 2026] Title:Reducing Text Bias in Synthetically Generated MCQAs for VLMs in Autonomous Driving View PDF HTML (experimenta
• X’s ‘For You’ algorithm nudges users toward conservative content. • Study surveyed nearly 5,000 X users across diverse demographics. • Political opinion shifts were measurable af
• AI increasingly shapes health and justice decisions, but bias remains a critical challenge. • UC Berkeley’s Emma Pierson develops fair algorithms to reduce racial and gender bias
• ChatGPT excels in English but favors Standard American/British over other dialects. • Study tested GPT‑3.5 Turbo and GPT‑4 across 10 English varieties. • Non‑standard dialects re
• Sample language model responses to different varieties of English and native speaker reactions. • ChatGPT does amazingly well at communicating with people in English. • Only 15%o