Pith. sign in

REVIEW 5 cited by

Ghostbuster: Detecting Text Ghostwritten by Large Language Models

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2305.15047 v3 pith:SPFMQKRG submitted 2023-05-24 cs.CL cs.AI

classification cs.CLcs.AI
keywords ghostbustermodelmodelstextai-generateddetectingdocumentsdomains
verification ladder T0 review T1 audit T2 compute T3 formal

Signed reviews

No signed human review yet.

0 comments
read the original abstract

We introduce Ghostbuster, a state-of-the-art system for detecting AI-generated text. Our method works by passing documents through a series of weaker language models, running a structured search over possible combinations of their features, and then training a classifier on the selected features to predict whether documents are AI-generated. Crucially, Ghostbuster does not require access to token probabilities from the target model, making it useful for detecting text generated by black-box models or unknown model versions. In conjunction with our model, we release three new datasets of human- and AI-generated text as detection benchmarks in the domains of student essays, creative writing, and news articles. We compare Ghostbuster to a variety of existing detectors, including DetectGPT and GPTZero, as well as a new RoBERTa baseline. Ghostbuster achieves 99.0 F1 when evaluated across domains, which is 5.9 F1 higher than the best preexisting model. It also outperforms all previous approaches in generalization across writing domains (+7.5 F1), prompting strategies (+2.1 F1), and language models (+4.4 F1). We also analyze the robustness of our system to a variety of perturbations and paraphrasing attacks and evaluate its performance on documents written by non-native English speakers.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 5 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Detecting LLM-generated Code with Subtle Modification by Adversarial Training

    cs.SE 2025-07 conditional novelty 6.0 of 10

    CodeGPTSensor+, trained with adversarial samples that combine identifier renaming and structure transformation, is substantially more robust to subtle modifications of LLM-generated code than the original CodeGPTSensor.

  2. DAMAGE: Detecting Adversarially Modified AI Generated Text

    cs.CL 2025-01 conditional novelty 6.0 of 10

    Adding humanizer-processed text to training data yields a detector that catches 98.26% of humanized AI essays at a 5% false-positive rate and stays robust to a detector-targeted attack.

  3. Who Gets Seen in the Age of AI? Adoption Patterns of Large Language Models in Scholarly Writing and Citation Outcomes

    cs.CY 2025-09 reject novelty 5.0 of 10

    Analyzing 98,000 Scopus computer science papers, the paper finds a global rise in AI-like writing after ChatGPT and reports regional differences in citation returns, but the key differential-gain result is statistical...

  4. GenAI Content Detection Task 1: English and Multilingual Machine-Generated Text Detection: AI vs. Human

    cs.CL 2025-01 conditional novelty 5.0 of 10

    A COLING 2025 shared task benchmark showing that current machine-generated text detectors reach only moderate accuracy and degrade badly on out-of-domain and humanized AI text.

  5. Towards Structurally Explainable Machine-Generated Text Detection: A Graph-Perspective Framework

    cs.CL 2025-05 reject novelty 4.0 of 10

    LM2OTIFS uses word co-occurrence graphs and GNNExplainer to detect and explain machine-generated text, with strong in-domain accuracy but unsupported faithfulness claims and a flawed theoretical proof.

Pith tools