Pith. sign in

REVIEW 3 cited by

DAMAGE: Detecting Adversarially Modified AI Generated Text

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 2501.03437 v1 pith:AWBYDBEE submitted 2025-01-06 cs.CL

Elyas Masrour , Bradley Emi , Max Spero This is my paper · ORCID
classification cs.CL
keywords textdetectorattackdetecthumanizedmodelrobustsoftware
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

AI humanizers are a new class of online software tools meant to paraphrase and rewrite AI-generated text in a way that allows them to evade AI detection software. We study 19 AI humanizer and paraphrasing tools and qualitatively assess their effects and faithfulness in preserving the meaning of the original text. We show that many existing AI detectors fail to detect humanized text. Finally, we demonstrate a robust model that can detect humanized AI text while maintaining a low false positive rate using a data-centric augmentation approach. We attack our own detector, training our own fine-tuned model optimized against our detector's predictions, and show that our detector's cross-humanizer generalization is sufficient to remain robust to this attack.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 3 Pith papers

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

  1. T5-CSBoost: Adversarial Perturbation Resistant LLM Fingerprinting

    cs.CL 2026-05 conditional novelty 6.0 of 10

    Adding a margin-based triplet loss to T5-Sentinel's decoder embeddings improves LLM source attribution robustness to word/character edits, paraphrasing, and unseen models/domains.

  2. Pangram 4 Technical Report

    cs.CL 2026-07 conditional novelty 5.0 of 10

    Pangram 4 is a commercial MoE-based detector claiming 0.9916 AUROC, 0.0041% FPR, 0.3396% FNR, plus tokenwise human/AI-assisted/AI-generated labels and humanizer detection.

  3. Can You Detect the Difference?

    cs.CL 2025-07 reject novelty 4.0 of 10

    A 2,000-sample comparison finds diffusion-generated LLaDA text can match human perplexity and burstiness when rephrasing, while LLaMA text is more predictable and easier to flag.

Pith tools