mlsanityData to 14 Oct 2026

Zero-Shot Detection of LLM-Generated Text via Implicit Reward Model

arXiv 2604.21223 ↗First posted 23 Apr 2026cs.CL cs.AIcited by 4

On the map: LLM alignment / RLHF / preference optimisation → Human values and AI feedback alignment

Abstract (arXiv)

Large language models (LLMs) have demonstrated remarkable capabilities across various tasks. However, their ability to generate human-like text has raised concerns about potential misuse. This underscores the need for reliable and effective methods to detect LLM-generated text. In this paper, we propose IRM, a novel zero-shot approach that leverages Implicit Reward Models for LLM-generated text detection. Such implicit reward models can be derived from publicly available instruction-tuned and base models. Previous reward-based method relies on preference construction and task-specific fine-tuning. In comparison, IRM requires neither preference collection nor additional training. We evaluate IRM on the DetectRL benchmark and demonstrate that IRM can achieve superior detection performance, outperforms existing zero-shot and supervised methods in LLM-generated text detection.

Key facts. Cited by 4 papers in the map (data through 14 Oct 2026). Placed in Human values and AI feedback alignment, one of 10 directions in LLM alignment / RLHF / preference optimisation. Builds on 28 of its 32 references that are in the map. First posted on arXiv 23 Apr 2026.

Builds on

28 of 32 references.

Closest ancestors · 2

Same direction, newest first

  1. 2024ReMoDetect: Reward Models Recognize Aligned LLM's Generations27 May
  2. 2020Learning to summarize from human feedback2 Sep

Other references · 26

  1. 2020Language Models are Few-Shot Learnerscited by 32,419
  2. 2022Training language models to follow instructions with human feedbackcited by 15,971
  3. 2023Direct Preference Optimization: Your Language Model is Secretly a Reward Modelcited by 8,065
  4. 2023DetectGPT: Zero-Shot Machine-Generated Text Detection using Probability Curvaturecited by 726
  5. 2019Defending Against Neural Fake Newscited by 559
  6. 2019Release Strategies and the Social Impacts of Language Modelscited by 513
  7. 2019GLTR: Statistical Detection and Visualization of Generated Textcited by 496
  8. 2023Fast-DetectGPT: Efficient Zero-Shot Detection of Machine-Generated Text via Conditional Probability Curvaturecited by 287
  9. 2024Spotting LLMs With Binoculars: Zero-Shot Detection of Machine-Generated Textcited by 263
  10. 2025Skywork-Reward-V2: Scaling Preference Data Curation via Human-AI Synergycited by 214
  11. 2023DetectLLM: Leveraging Log Rank Information for Zero-Shot Detection of Machine-Generated Textcited by 210
  12. 2023RADAR: Robust AI-Text Detection via Adversarial Learningcited by 190
14 more
  1. 2024Free Process Rewards without Process Labelscited by 165
  2. 2024RAID: A Shared Benchmark for Robust Evaluation of Machine-Generated Text Detectorscited by 163
  3. 2024Regularizing Hidden States Enables Learning Generalizable Reward Model for LLMscited by 142
  4. 2023DNA-GPT: Divergent N-Gram Analysis for Training-Free Detection of GPT-Generated Textcited by 129
  5. 2019Unifying Human and Statistical Evaluation for Natural Language Generationcited by 127
  6. 2019Real or Fake? Learning to Discriminate Machine from Human Generated Textcited by 99
  7. 2023Ghostbuster: Detecting Text Ghostwritten by Large Language Modelscited by 91
  8. 2024Raidar: geneRative AI Detection viA Rewritingcited by 52
  9. 2024DetectRL: Benchmarking LLM-Generated Text Detection in Real-World Scenarioscited by 38
  10. 2024Training-free LLM-generated Text Detection by Mining Token Probability Sequencescited by 35
  11. 2024Detecting Subtle Differences between Human and Model Languages Using Spectrum of Relative Likelihoodcited by 24
  12. 2024Glimpse: Enabling White-Box Methods to Use Proprietary Models for Zero-Shot LLM-Generated Text Detectioncited by 14
  13. 2024Zero-Shot Detection of LLM-Generated Text using Token Cohesivenesscited by 13
  14. 2025DivScore: Zero-Shot Detection of LLM-Generated Text in Specialized Domainscited by 9

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 4.

4 papers.

  1. 2026DetectZoo: A Unified Toolkit for AI-Generated Content Detection Across Text, Audio, and Image Modalitiescited by 1
  2. 2026Once a Response, Always a Response: Detecting LLM-generated Text via Latent Prompt Restorationcited by 1
  3. 2026Steer-to-Detect: Probing Hidden Representations for Detection of LLM-Generated Textscited by 0
  4. 2026Who's Behind the Text? Robust Model Authorship Attributioncited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2023A Survey on LLM-Generated Text Detection: Necessity, Methods, and Future Directions
  2. 2025Understanding the Effects of RLHF on the Quality and Detectability of LLM-Generated Texts
  3. 2024Understanding the Effects of Human-written Paraphrases in LLM-generated Text Detection
  4. 2023From Text to Source: Results in Detecting Large Language Model-Generated Content
  5. 2026Detecting the Machine: A Comprehensive Benchmark of AI-Generated Text Detectors Across Architectures, Domains, and Adversarial Conditions
  6. 2025Kill two birds with one stone: generalized and robust AI-generated text detection via dynamic perturbations
  7. 2024A Critical Look At Tokenwise Reward-Guided Text Generation
  8. 2025RepreGuard: Detecting LLM-Generated Text by Revealing Hidden Representation Patterns