Pith. sign in

REVIEW 11 cited by

Fast-DetectGPT: Efficient Zero-Shot Detection of Machine-Generated Text via Conditional Probability Curvature

Not yet reviewed by Pith; the record is open.

Re-run · record.json Download PDF Read on arXiv ↗

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 2310.05130 v3 pith:WVELVWN4 submitted 2023-10-08 cs.CL

Guangsheng Bao , Yanbin Zhao , Zhiyang Teng , Linyi Yang , Yue Zhang This is my paper
classification cs.CL
keywords curvaturedetectgptfast-detectgptzero-shotconditionalcontentdetectiondetector
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

Large language models (LLMs) have shown the ability to produce fluent and cogent content, presenting both productivity opportunities and societal risks. To build trustworthy AI systems, it is imperative to distinguish between machine-generated and human-authored content. The leading zero-shot detector, DetectGPT, showcases commendable performance but is marred by its intensive computational costs. In this paper, we introduce the concept of conditional probability curvature to elucidate discrepancies in word choices between LLMs and humans within a given context. Utilizing this curvature as a foundational metric, we present **Fast-DetectGPT**, an optimized zero-shot detector, which substitutes DetectGPT's perturbation step with a more efficient sampling step. Our evaluations on various datasets, source models, and test conditions indicate that Fast-DetectGPT not only surpasses DetectGPT by a relative around 75% in both the white-box and black-box settings but also accelerates the detection process by a factor of 340, as detailed in Table 1. See \url{https://github.com/baoguangsheng/fast-detect-gpt} for code, data, and results.

Discussion (0). Sign in to comment.

Forward citations

Cited by 11 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. OpenAlex reports about 19 citations worldwide. Full citation record

  1. HIPPO-Video: Simulating Watch Histories with Large Language Models for Personalized Video Highlighting

    cs.CV 2025-07 conditional novelty 7.0 of 10

    HIPPO-Video contributes 2,040 LLM-simulated watch-history and saliency-score pairs, and the HiPHer model uses these histories to beat generic and query-based baselines on the new benchmark.

  2. Zero-Shot Detection of LLM-Generated Code via Approximated Task Conditioning

    cs.CL 2025-06 conditional novelty 7.0 of 10

    ATC detects AI-generated code by asking a language model to reconstruct the programming task, then scoring token entropy under that reconstructed task, outperforming prior zero-shot detectors on Python, C++, and Java ...

  3. When Words Predict Workload

    cs.DC 2026-07 conditional novelty 6.0 of 10

    Linguistic features plus a closed-form latency threshold route patent-claim LLM requests before edge GPU allocation, cutting misroutes from 0.85 to ~0.09 while bounding VRAM at 4.82 GiB.

  4. AEyeDE: An Attention-Based Attribution Framework for AI-Generated Text Detection

    cs.CL 2026-04 conditional novelty 6.0 of 10

    Attention attribution maps from a white-box proxy Transformer, classified by a lightweight CNN, provide a competitive and interpretable signal for AI-generated text detection.

  5. MAGA-Bench: Machine-Augment-Generated Text via Alignment Detection Benchmark

    cs.CL 2026-01 conditional novelty 6.0 of 10

    Adding human-alignment augmentation (roleplaying, BPO, self-refine, RLDF) to machine-generated text both fools existing detectors and improves the generalization of detectors fine-tuned on it.

  6. Benchmarking the Detection of LLMs-Generated Modern Chinese Poetry

    cs.CL 2025-09 conditional novelty 6.0 of 10

    A new modern Chinese poetry detection benchmark shows most current AI-text detectors are unreliable, particularly when LLMs imitate a human style.

  7. Beyond Easy Wins: A Text Hardness-Aware Benchmark for LLM-generated Text Detection

    cs.CL 2025-07 conditional novelty 6.0 of 10

    SHIELD shows that standard AUROC overstates AI-text detector quality, and that six zero-shot detectors collapse under a controllable word-replacement humanification.

  8. Zero-Shot Attribution for Large Language Models: A Distribution Testing Approach

    cs.LG 2025-06 conditional novelty 6.0 of 10

    Anubis re-frames LLM attribution as a distribution testing problem with EVAL access, and reports AUROC above 0.9 on code benchmarks with around 2000 samples, beating detectGPT.

  9. Frontier AI Risk Management Framework in Practice: A Risk Analysis Technical Report

    cs.AI 2025-07 conditional novelty 5.0 of 10

    An evaluation of 18 frontier AI models across seven catastrophic-risk categories finds all models in green or yellow zones, with none crossing the report's proposed red lines.

  10. A Comprehensive Dataset for Human vs. AI Generated Text Detection

    cs.CL 2025-10 reject novelty 4.0 of 10

    A dataset of ~58k NYT articles plus AI rewrites from six LLMs, evaluated with a rewrite-distance baseline reaching 58.35% detection and 8.92% attribution accuracy.

  11. Efficient Online LLM Watermark Detection via Rao-Blackwellized E-Processes

    stat.ML 2026-07 conditional novelty 3.0 of 10

    A one-pass e-process detector for the Gumbel-max LLM watermark preserves anytime-valid Type I error control while accumulating evidence token by token.

Pith tools