Semantic Reconstruction of Adversarial Plagiarism: A Context-Aware Framework for Detecting and Restoring "Tortured Phrases" in Scientific Literature

TL;DR

This paper introduces SRAP, a novel framework combining anomaly detection and semantic reconstruction to identify and restore paraphrased scientific text, significantly improving detection accuracy over existing methods.

Contribution

It presents a two-stage approach using domain-specific language models and dense retrieval for detecting and restoring adversarial plagiarism in scientific literature.

Findings

Achieves 23.67% restoration accuracy, outperforming baseline methods.

Static decision boundaries are more robust than dynamic thresholds in scientific jargon.

Zero-shot baselines fail completely in restoring obfuscated phrases.

Abstract

The integrity and reliability of scientific literature is facing a serious threat by adversarial text generation techniques, specifically from the use of automated paraphrasing tools to mask plagiarism. These tools generate "tortured phrases", statistically improbable synonyms (e.g. "counterfeit consciousness" for "artificial intelligence"), that preserve the local grammar while obscuring the original source. Most existing detection methods depend heavily on static blocklists or general-domain language models, which suffer from high false-negative rates for novel obfuscations and cannot determine the source of the plagiarized content. In this paper, we propose Semantic Reconstruction of Adversarial Plagiarism (SRAP), a framework designed not only to detect these anomalies but to mathematically recover the original terminology. We use a two-stage architecture: (1) statistical anomaly detection with a domain-specific masked language model (SciBERT) using token-level pseudo-perplexity, and (2) source-based semantic reconstruction using dense vector retrieval (FAISS) and sentence-level alignment (SBERT). Experiments on a parallel corpus of adversarial scientific text show that while zero-shot baselines fail completely (0.00 percent restoration accuracy), our retrieval-augmented approach achieves 23.67 percent restoration accuracy, significantly outperforming baseline methods. We also show that static decision boundaries are necessary for robust detection in jargon-heavy scientific text, since dynamic thresholding fails under high variance. SRAP enables forensic analysis by linking obfuscated expressions back to their most probable source documents.