Science Innovations: Natural Rainbow Colours Produced  |  Technology Inceptions: Muscope, World’s Smallest Microscope  |  Science Innovations: Ultrasensitive Tactile Sensors for Robots  |  Policy Indications: How Materials Science Helps Contain Contain Covid-19 Spread  |  National Edu News: IIT Hyderabad and PharmCADD signed a pact for the co-development of new drugs   |  Teacher Insights: Be Game  |  Health Monitor: Understanding ‘Haemorrhage'  |  National Edu News: Pallikkutam GlobalConnect#3 on 'Innovative Tools for Effective Teaching'  |  Expert Counsel: The Nine Dash Line  |  National Edu News: Astronomers Find One Group of Appearing and Disappearing Stars  |  Teacher Insights: Bird Book for Children to Love Nature  |  International Edu News: New Model to Fight Social Media Deep Fakes  |  Teacher Insights: Universal Lunch Makes Students Healthier  |  Teacher Insights: Physical Activity Boosts Self Regulation  |  Parent Interventions: Anti-Inflammatory Foods Reduce Blood Fats  |  
July 27, 2019 Saturday 09:51:27 AM IST

Statistical Method Identified to Detect Fake or Computer-Generated Text

By Felix Litchenfeld

Researchers at Harvard John A Paulson School of Engineering and Applied Sciences (SEAS) and IBM have developed a statistical method to detect computer-generated or fake text from human generated text.
Researchers Sebastian Gehrmann and Hendrik Strobert(IBM) found that natural-language generators are trained on tens of millions of online texts and mimic human language by predicting the words that most often follow one another. For eg I followed by 'have' and 'am'. Using this idea they developed a method that identifies predictable text instead of flagging errors. Gehrmann and Strobelt’s method, known as GLTR, is based on a model trained on 45 million texts from websites — the public version of the OpenAI model, GPT-2. Because it uses GPT-2 to detect generated text, GLTR works best against GPT-2, but it does well against other models, too.


https://news.harvard.edu/gazette/story/2019/07/researchers-develop-a-method-to-identify-computer-generated-text/?utm_medium=Feed&utm_source=Syndication



Comments