mir/gpt-2-output-dataset
2
0
Fork 0
mirror of https://github.com/openai/gpt-2-output-dataset synced 2025-08-22 18:07:53 +00:00
Code Issues Packages Projects Releases Wiki Activity

tweak size

Browse Source
This commit is contained in:
Alec 2019-05-03 16:10:02 -07:00
parent 3502958089
commit 2ad1f71e23
1 changed files with 2 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
README.md
View File

@ -43,11 +43,11 @@ We've provided a starter baseline which trains a logistic regression detector on
### Initial Analysis ### Initial Analysis
<img src="https://i.imgur.com/PZ3GOeS.png" width="713" height="502" title="Impact of Document Length"> <img src="https://i.imgur.com/PZ3GOeS.png" width="475" height="335" title="Impact of Document Length">
Shorter documents are harder to detect. Accuracy of detection of a short documents of 500 characters (a long paragraph) is about 15% lower. Shorter documents are harder to detect. Accuracy of detection of a short documents of 500 characters (a long paragraph) is about 15% lower.
<img src="https://i.imgur.com/eH9Ogqo.png" width="723" height="450" title="Part of Speech Analysis"> <img src="https://i.imgur.com/eH9Ogqo.png" width="482" height="300" title="Part of Speech Analysis">
Truncated sampling, which is commonly used for high-quality generations from the GPT-2 model family, results in a shift in the part of speech distribution of the generated text compared to real text. A clear example is the underuse of proper nouns and overuse of pronouns which are more generic. This shift contributes to the 8% to 18% higher detection rate of Top-K samples compared to random samples across models. Truncated sampling, which is commonly used for high-quality generations from the GPT-2 model family, results in a shift in the part of speech distribution of the generated text compared to real text. A clear example is the underuse of proper nouns and overuse of pronouns which are more generic. This shift contributes to the 8% to 18% higher detection rate of Top-K samples compared to random samples across models.