vminvsky

4 ms·

1. 1.▲
   AI-assisted reviews lead to higher chance of acceptance(twitter.com)

   1 points by vminvsky 2y ago|0 comments
2. 2.▲
   Using Negotiating Agents to Evaluate LLMs(twitter.com)

   2 points by vminvsky 3y ago|0 comments
3. 3.▲
   by vminvsky 3y ago

   They never used a static detector.
4. 4.▲
   by vminvsky 3y ago

   i would recommend u read the paper. the contribution isnt a detector thats meant to be taken seriously; but a detector that works in a very specific task. they then use this to estimate use of LLMs on MTurk
5. 5.▲
   by vminvsky 3y ago

   according to the paper they get 98% accuracy. another recent paper came out saying it's always possible to discriminate between real and synthetic text [1]. i think the core problem is with the generalist classifiers (gptzero, openai d
6. 6.▲
   by vminvsky 3y ago

   It's actually pretty easy to create a bespoke ChatGPT detector!
7. 7.▲
   Show HN: ChatGPT for Student Courses(pdf.asknostoi.com)

   2 points by vminvsky 4y ago|0 comments
8. 8.▲
   by vminvsky 4y ago

   What about double-descent :D