AI Snake Oil personal blog
GPT-4 and professional benchmarks: the wrong answer to the wrong question
OpenAI may have tested on the training data. Besides, human benchmarks are meaningless for bots.
- Points
- 340
- HN comments
- 185
- Submissions
- 1
- Cited by
- 4 commenters
AI & Machine LearningScience & Space2023
On Hacker News
| Submitted | Title on HN | Points | Comments |
|---|---|---|---|
| Mar 21, 2023 | GPT-4 and professional benchmarks: the wrong answer to the wrong question by agomez314 | 340 | 185 |
Cited in comments
Linked from 5 Hacker News comments by 4 different people, from Mar 2023 to Apr 2023.
Most recent:
- civilized · Apr 13, 2023
- godelski · Apr 8, 2023
- avianlyric · Apr 4, 2023
- civilized · Mar 26, 2023
- kgarten · Mar 21, 2023
