Cleansing Jewel: A Neural Spelling Correction Model Built On Google OCR-ed Tibetan Manuscripts

no code implementations • 7 Apr 2023 • Queenie Luo, Yung-Sung Chuang

Then, we implemented a Confidence Score mechanism into the Transformer architecture to perform spelling correction tasks.

Optical Character Recognition Optical Character Recognition (OCR) +1

A "Perspectival" Mirror of the Elephant: Investigating Language Bias on Google, ChatGPT, YouTube, and Wikipedia

no code implementations • 28 Mar 2023 • Queenie Luo, Michael J. Puett, Michael D. Smith

Contrary to Google Search's mission of delivering information from "many angles so you can form your own understanding of the world," we find that Google and its most prominent returned results - Wikipedia and YouTube - simply reflect a narrow set of culturally dominant views tied to the search language for complex topics like "Buddhism," "Liberalism," "colonization," "Iran" and "America."

Cultural Vocal Bursts Intensity Prediction