Tuning for TraceTarnish: Techniques, Trends, and Testing Tangible Traits

• URL: http://arxiv.org/abs/2512.03465v1
• Date: Wed, 03 Dec 2025 05:39:40 GMT
• Title: Tuning for TraceTarnish: Techniques, Trends, and Testing Tangible Traits
• Authors: Robert Dilworth,
• Abstract summary: Attack script $textitTraceTarnish$ uses adversarial stylometry principles to anonymize the authorship of text-based messages.<n>Stylistometric cues--function-word frequencies, content-word distributions, and the Type-Token Ratio--serve as reliable indicators of compromise.
• Score: 0.0
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: In this study, we more rigorously evaluated our attack script $\textit{TraceTarnish}$, which leverages adversarial stylometry principles to anonymize the authorship of text-based messages. To ensure the efficacy and utility of our attack, we sourced, processed, and analyzed Reddit comments--comments that were later alchemized into $\textit{TraceTarnish}$ data--to gain valuable insights. The transformed $\textit{TraceTarnish}$ data was then further augmented by $\textit{StyloMetrix}$ to manufacture stylometric features--features that were culled using the Information Gain criterion, leaving only the most informative, predictive, and discriminative ones. Our results found that function words and function word types ($L\_FUNC\_A$ $\&$ $L\_FUNC\_T$); content words and content word types ($L\_CONT\_A$ $\&$ $L\_CONT\_T$); and the Type-Token Ratio ($ST\_TYPE\_TOKEN\_RATIO\_LEMMAS$) yielded significant Information-Gain readings. The identified stylometric cues--function-word frequencies, content-word distributions, and the Type-Token Ratio--serve as reliable indicators of compromise (IoCs), revealing when a text has been deliberately altered to mask its true author. Similarly, these features could function as forensic beacons, alerting defenders to the presence of an adversarial stylometry attack; granted, in the absence of the original message, this signal may go largely unnoticed, as it appears to depend on a pre- and post-transformation comparison. "In trying to erase a trace, you often imprint a larger one." Armed with this understanding, we framed $\textit{TraceTarnish}$'s operations and outputs around these five isolated features, using them to conceptualize and implement enhancements that further strengthen the attack.

Related papers

• StegoStylo: Squelching Stylometric Scrutiny through Steganographic Stitching [0.0]
  Stylometry supports copyright and plagiarism investigations, aids detection of harmful content, and provides historical context for literary works.<n>Stylometry is employed as a tool for authorship verification--confirming whether a text truly originates from a claimed author--it can also be weaponized for malicious purposes.<n>This paper explores how adversarial stylometry combined with steganography can counteract stylometric analysis.
  arXiv  Detail & Related papers  (2026-01-14T00:49:20Z)
• Feedback Descent: Open-Ended Text Optimization via Pairwise Comparison [48.89195616081196]
  Feedback Descent is a framework that optimize text artifacts -- prompts, code, and molecules -- through structured textual feedback.<n>We show that in-context learning can transform structured feedback into gradient-like directional information, enabling targeted edits.<n>In the DOCKSTRING molecule discovery benchmark, Feedback Descent identifies novel drug-like molecules surpassing the $99.9$th percentile of a database with more than $260,000$ compounds across six protein targets.
  arXiv  Detail & Related papers  (2025-11-11T07:14:13Z)
• When Personalization Tricks Detectors: The Feature-Inversion Trap in Machine-Generated Text Detection [64.23509202768945]
  We introduce dataset, the first benchmark for evaluating detector robustness in personalized settings.<n>Our experimental results demonstrate large performance gaps across detectors in personalized settings.<n>We propose method, a simple and reliable way to predict detector performance changes in personalized settings.
  arXiv  Detail & Related papers  (2025-10-14T13:10:23Z)
• Diversity Boosts AI-Generated Text Detection [51.56484100374058]
  DivEye is a novel framework that captures how unpredictability fluctuates across a text using surprisal-based features.<n>Our method outperforms existing zero-shot detectors by up to 33.2% and achieves competitive performance with fine-tuned baselines.
  arXiv  Detail & Related papers  (2025-09-23T10:21:22Z)
• Towards Bridging Review Sparsity in Recommendation with Textual Edge Graph Representation [28.893058826607735]
  We propose a unified framework that imputes missing reviews by jointly modeling semantic and structural signals.<n>Experiments on the Amazon and Goodreads datasets show that TWISTER consistently outperforms traditional numeric, graph-based, and LLM baselines.<n>In summary, TWISTER generates reviews that are more helpful, authentic, and specific, while smoothing structural signals for improved recommendations.
  arXiv  Detail & Related papers  (2025-08-02T00:53:40Z)
• Towards Generalized and Training-Free Text-Guided Semantic Manipulation [123.80467566483038]
  Text-guided semantic manipulation refers to semantically editing an image generated from a source prompt to match a target prompt.<n>We propose a novel $textitGTF$ for text-guided semantic manipulation, which has the following attractive capabilities.<n>Our experiments demonstrate the efficacy of our approach, highlighting its potential to advance the state-of-the-art in semantics manipulation.
  arXiv  Detail & Related papers  (2025-04-24T05:54:56Z)
• Breaking BERT: Gradient Attack on Twitter Sentiment Analysis for Targeted Misclassification [0.0]
  Bidirectional Representations from Transformers BERT has been widely adapted in sentiment analysis.<n>BERT is susceptible to adversarial attacks.<n>This paper aims to scrutinize the inherent vulnerabilities of such models in Twitter sentiment analysis.
  arXiv  Detail & Related papers  (2025-04-02T04:21:19Z)
• Discourse Features Enhance Detection of Document-Level Machine-Generated Content [53.41994768824785]
  Machine-generated content poses challenges such as academic plagiarism and the spread of misinformation.<n>Existing MGC detectors often focus solely on surface-level information, overlooking implicit and structural features.<n>We introduce novel methodologies and datasets to overcome these challenges.
  arXiv  Detail & Related papers  (2024-12-17T08:47:41Z)
• Integrating Bidirectional Long Short-Term Memory with Subword Embedding for Authorship Attribution [2.3429306644730854]
  Manifold word-based stylistic markers have been successfully used in deep learning methods to deal with the intrinsic problem of authorship attribution. The proposed method was experimentally evaluated against numerous state-of-the-art methods across the public corporal of CCAT50, IMDb62, Blog50, and Twitter50.
  arXiv  Detail & Related papers  (2023-06-26T11:35:47Z)
• Scaling up sign spotting through sign language dictionaries [99.50956498009094]
  The focus of this work is $textitsign spotting$ - given a video of an isolated sign, our task is to identify $textitwhether$ and $textitwhere$ it has been signed in a continuous, co-articulated sign language video. We train a model using multiple types of available supervision by: (1) $textitwatching$ existing footage which is sparsely labelled using mouthing cues; (2) $textitreading$ associated subtitles which provide additional translations of the signed content. We validate the effectiveness of our approach on low
  arXiv  Detail & Related papers  (2022-05-09T10:00:03Z)
• Towards Variable-Length Textual Adversarial Attacks [68.27995111870712]
  It is non-trivial to conduct textual adversarial attacks on natural language processing tasks due to the discreteness of data. In this paper, we propose variable-length textual adversarial attacks(VL-Attack) Our method can achieve $33.18$ BLEU score on IWSLT14 German-English translation, achieving an improvement of $1.47$ over the baseline model.
  arXiv  Detail & Related papers  (2021-04-16T14:37:27Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.