61 lines
2.2 KiB
Markdown
61 lines
2.2 KiB
Markdown
---
|
|
date: '2023-01-29T10:35:56'
|
|
hypothesis-meta:
|
|
created: '2023-01-29T10:35:56.649264+00:00'
|
|
document:
|
|
title:
|
|
- 2301.11305.pdf
|
|
flagged: false
|
|
group: __world__
|
|
hidden: false
|
|
id: tr0lTp_AEe2k81d5ilJ0Xw
|
|
links:
|
|
html: https://hypothes.is/a/tr0lTp_AEe2k81d5ilJ0Xw
|
|
incontext: https://hyp.is/tr0lTp_AEe2k81d5ilJ0Xw/arxiv.org/pdf/2301.11305.pdf
|
|
json: https://hypothes.is/api/annotations/tr0lTp_AEe2k81d5ilJ0Xw
|
|
permissions:
|
|
admin:
|
|
- acct:ravenscroftj@hypothes.is
|
|
delete:
|
|
- acct:ravenscroftj@hypothes.is
|
|
read:
|
|
- group:__world__
|
|
update:
|
|
- acct:ravenscroftj@hypothes.is
|
|
tags:
|
|
- chatgpt
|
|
- detecting gpt
|
|
target:
|
|
- selector:
|
|
- end: 1096
|
|
start: 756
|
|
type: TextPositionSelector
|
|
- exact: his approach, which we call DetectGPT,does not require training a separate
|
|
classifier, col-lecting a dataset of real or generated passages, orexplicitly
|
|
watermarking generated text. It usesonly log probabilities computed by the
|
|
model ofinterest and random perturbations of the passagefrom another generic
|
|
pre-trained language model(e.g, T5)
|
|
prefix: ' is generated from a givenLLM. T'
|
|
suffix: . We find DetectGPT is more disc
|
|
type: TextQuoteSelector
|
|
source: https://arxiv.org/pdf/2301.11305.pdf
|
|
text: The novelty of this approach is that it is cheap to set up as long as you
|
|
have the log probabilities generated by the model of interest.
|
|
updated: '2023-01-29T10:35:56.649264+00:00'
|
|
uri: https://arxiv.org/pdf/2301.11305.pdf
|
|
user: acct:ravenscroftj@hypothes.is
|
|
user_info:
|
|
display_name: James Ravenscroft
|
|
in-reply-to: https://arxiv.org/pdf/2301.11305.pdf
|
|
tags:
|
|
- chatgpt
|
|
- detecting gpt
|
|
- hypothesis
|
|
type: annotation
|
|
url: /annotations/2023/01/29/1674988556
|
|
|
|
---
|
|
|
|
|
|
|
|
<blockquote>his approach, which we call DetectGPT,does not require training a separate classifier, col-lecting a dataset of real or generated passages, orexplicitly watermarking generated text. It usesonly log probabilities computed by the model ofinterest and random perturbations of the passagefrom another generic pre-trained language model(e.g, T5)</blockquote>The novelty of this approach is that it is cheap to set up as long as you have the log probabilities generated by the model of interest. |