mlsanityData to 14 Oct 2026

Copyright Detective: A Forensic System to Evidence LLMs Flickering Copyright Leakage Risks

arXiv 2602.05252 ↗First posted 5 Feb 2026cs.CLcited by 1

On the map: LLM safety and security → Watermarking generated content and models

Abstract (arXiv)

We present Copyright Detective, the first interactive forensic system for detecting, analyzing, and visualizing potential copyright risks in LLM outputs. The system treats copyright infringement versus compliance as an evidence discovery process rather than a static classification task due to the complex nature of copyright law. It integrates multiple detection paradigms, including content recall testing, paraphrase-level similarity analysis, persuasive jailbreak probing, and unlearning verification, within a unified and extensible framework. Through interactive prompting, response collection, and iterative workflows, our system enables systematic auditing of verbatim memorization and paraphrase-level leakage, supporting responsible deployment and transparent evaluation of LLM copyright risks even with black-box access.

Key facts. Cited by 1 paper in the map (data through 14 Oct 2026). Placed in Watermarking generated content and models, one of 12 directions in LLM safety and security. Builds on 28 of its 45 references that are in the map. First posted on arXiv 5 Feb 2026.

Builds on

28 of 45 references.

References · 28

  1. 2020The Pile: An 800GB Dataset of Diverse Text for Language Modelingcited by 2,411
  2. 2019Similarity of Neural Network Representations Revisitedcited by 1,755
  3. 2023Can LLMs Express Their Uncertainty? An Empirical Evaluation of Confidence Elicitation in LLMscited by 974
  4. 2024TOFU: A Task of Fictitious Unlearning for LLMscited by 533
  5. 2024Negative Preference Optimization: From Catastrophic Collapse to Effective Unlearningcited by 504
  6. 2024MUSE: Machine Unlearning Six-Way Evaluation for Language Modelscited by 261
  7. 2023Large Language Model Alignment: A Surveycited by 243
  8. 2023Speak, Memory: An Archaeology of Books Known to ChatGPT/GPT-4cited by 125
  9. 2023Quantifying Uncertainty in Answers from any Language Model and Enhancing their Trustworthinesscited by 122
  10. 2024LUQ: Long-text Uncertainty Quantification for LLMscited by 81
  11. 2024Catastrophic Failure of LLM Unlearning via Quantizationcited by 81
  12. 2024DE-COP: Detecting Copyrighted Content in Language Models Training Datacited by 74
16 more
  1. 2022Do Language Models Plagiarize?cited by 73
  2. 2024Operationalizing a Threat Model for Red-Teaming Large Language Models (LLMs)cited by 46
  3. 2025Unlearning Isn't Deletion: Investigating Reversibility of Machine Unlearning in LLMscited by 42
  4. 2024CopyBench: Measuring Literal and Non-Literal Reproduction of Copyright-Protected Text in Language Model Generationcited by 35
  5. 2025Extracting memorized pieces of (copyrighted) books from open-weight language modelscited by 30
  6. 2025GUARD: Generation-time LLM Unlearning via Adaptive Restriction and Detectioncited by 17
  7. 2026Extracting books from production language modelscited by 17
  8. 2024Evaluating Uncertainty-based Failure Detection for Closed-Loop LLM Plannerscited by 15
  9. 2025An overview of model uncertainty and variability in LLM-based sentiment analysis. Challenges, mitigation strategies and the role of explainabilitycited by 15
  10. 2024Do LLMs Know to Respect Copyright Notice?cited by 13
  11. 2024Measuring Copyright Risks of Large Language Model via Partial Information Probingcited by 9
  12. 2025Step-by-Step Reasoning Attack: Revealing 'Erased' Knowledge in Large Language Modelscited by 6
  13. 2025SUA: Stealthy Multimodal Large Language Model Unlearning Attackcited by 4
  14. 2025GRAIL: Gradient-Based Adaptive Unlearning for Privacy and Copyright in LLMscited by 3
  15. 2025DP2Unlearning: An Efficient and Guaranteed Unlearning Framework for LLMscited by 2
  16. 2025ISACL: Internal State Analyzer for Copyrighted Training Data Leakagecited by 2

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 1.

1 paper.

  1. 2026LLM-Oriented Information Retrieval: A Denoising-First Perspectivecited by 1

Similar, not cited

Close in wording, no citation either way

  1. 2025SoK: Large Language Model Copyright Auditing via Fingerprinting
  2. 2026AI Watermark Evidence Fails Forensic Readiness: An Empirical Evaluation
  3. 2025CopyJudge: Automated Copyright Infringement Identification and Mitigation in Text-to-Image Diffusion Models
  4. 2025PromptCOS: Towards Content-only System Prompt Copyright Auditing for LLMs
  5. 2025Copyright Protection for Large Language Models: A Survey of Methods, Challenges, and Trends
  6. 2026Tracing Provenance and Detecting Tampering with Complementary LLM Watermarks
  7. 2026DCS: A Unified Conditional Sensitivity Framework for Cross-Modal Copyright Infringement Detection
  8. 2026CopyShield: A Cross-Level Benchmark of Copyright Defenses in LLMs