---
date: '2023-01-29T11:01:22'
hypothesis-meta:
  created: '2023-01-29T11:01:22.509728+00:00'
  document:
    title:
    - 2301.11305.pdf
  flagged: false
  group: __world__
  hidden: false
  id: RDbfNJ_EEe258oPTYQZGNA
  links:
    html: https://hypothes.is/a/RDbfNJ_EEe258oPTYQZGNA
    incontext: https://hyp.is/RDbfNJ_EEe258oPTYQZGNA/arxiv.org/pdf/2301.11305.pdf
    json: https://hypothes.is/api/annotations/RDbfNJ_EEe258oPTYQZGNA
  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: 11561
      start: 11236
      type: TextPositionSelector
    - exact: "As in prior work, we study a \u2018white box\u2019 setting (Gehrmannet\
        \ al., 2019) in which the detector may evaluate the log prob-ability of a\
        \ sample log p\u03B8 (x). The white box setting doesnot assume access to the\
        \ model architecture or parameters.While most public APIs for LLMs (such as\
        \ GPT-3) enablescoring text, some exceptions exist"
      prefix: ed samples to perform detection.
      suffix: . While most of our ex-periments
      type: TextQuoteSelector
    source: https://arxiv.org/pdf/2301.11305.pdf
  text: The authors assume white-box access to the log probability of a sample \(log
    p_{\Theta}(x)\) but do not require access to the model's actual architecture or
    weights.
  updated: '2023-01-29T11:01:22.509728+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/2023/01/29/1674990082

---



 <blockquote>As in prior work, we study a ‘white box’ setting (Gehrmannet al., 2019) in which the detector may evaluate the log prob-ability of a sample log pθ (x). The white box setting doesnot assume access to the model architecture or parameters.While most public APIs for LLMs (such as GPT-3) enablescoring text, some exceptions exist</blockquote>The authors assume white-box access to the log probability of a sample \(log p_{\Theta}(x)\) but do not require access to the model's actual architecture or weights.