Pangram 4 Technical Report
pangram-public.s3.us-east-1.amazonaws.com · 8,618 words · saved by 2 readers
N/A
Pangram 4 Technical Report Ben Glickenhaus * Katherine Thai * Jenna Russell Elyas Masrour Yue Han Max Spero Bradley Emi Pangram Labs University of Maryland Abstract We present Pangram 4, the latest deep-learning-based AI-text classification model from Pangram Labs. We achieve an AUROC of 0.9916 with a false positive rate of 0.0041% and a false negative rate of 0.3396%. In addition to its increased overall accuracy compared with Pangram 3, Pangram 4 exhibits superior out-of-distribution generalization and robustness to…
saved by
related reading
- Pangram 4 Technical Overviewpangram.com
- I don't care (anymore) if your writing got a Pangram false positiveandrewwu.substack.com
- Seeing in Pangram Space | Pangrampangram.com
- How does Pangram work?pangram.substack.com
- How does Pangram work?substack.com
- I Wouldn't Say Pangram is Broken, But I Would Say That It's Brittlesubstack.com
- GPT-4openai.com
- Why Perplexity and Burstiness Fail to Detect AI | Pangram Labspangram.com
- Ghostbuster: Detecting Text Ghostwritten by Large Language Modelsarxiv.org
- God help me am I about to change my view on AI-assisted writing?rivalvoices.substack.com
- Employee Spotlight: Meet Katherine, AI Research Scientist | Pangram Labspangram.com
- gpt-4.pdfcdn.openai.com