Adnan KhanAdnanProduct & AI
San Francisco Bay Area
Get in touch

Source note2 min read

Detecting LLM-Generated Tokens in Human--LLM Coauthored Text

Source

arXiv

arxiv.org/abs/2607.21458v1

Read the source

Source

Plain-Language Brief

The paper presents a new method to detect which parts of a text are generated by language models (LLMs) in collaborative human-AI writing, operating at the token level and improving upon existing token-level detection scores. The method uses an adaptive Lepski-type rule to estimate the underlying signal and achieves favorable mean square error performance.

What the Document Says

  • The paper discusses a method for detecting parts of a text generated by language models (LLMs) in human-AI collaborative writing.
  • The method operates at the token level and uses existing token-level detection scores.
  • It smooths adjacent token scores to reduce variability.
  • An adaptive Lepski-type rule is used to select the bandwidth according to the local authorship structure.
  • The method does not require token-level labeled data for training.
  • Theoretically, it characterizes a trade-off and shows favorable mean square error performance.
  • Empirically, it demonstrates strong performance against various baselines in both synthetic datasets and realistic ones.

Key Concepts Mentioned

  • human-AI collaborative writing
  • language models
  • token-level detection
  • Lepski-type rule
  • mean square error

Claims Made by the Document

  • The method can accurately detect LLM-generated content in mixed-authorship documents.
  • The method achieves favorable mean square error performance.

Evidence Used

  • The paper demonstrates strong performance of the method against a wide range of baselines in both synthetic datasets and a realistic dataset.
  • Theoretically, it characterizes a trade-off and shows favorable mean square error performance.

Methods / Approach

UNKNOWN

Limitations Stated by the Authors

UNKNOWN

Limitations Noted by the Agent

Agent interpretation — not from the source.

UNKNOWN

Open Questions

UNKNOWN

Notes for Future Q&A

May include interpretation. Labeled where it does.

UNKNOWN

Raw Abstract / Excerpt

The rise of human-AI collaborative writing has created a growing need for fine-grained detection methods that support localizing likely LLM-generated content in mixed-authorship documents. Existing methods for detecting LLM-generated text mainly focus on document-level classification and cannot identify which parts of the text are generated by LLMs. This paper introduces a new method to address this urgent need. Our method operates at the token level, the natural unit of modern language models, and builds on existing token-level detection scores. The key idea is to smooth adjacent token scores to reduce their variability, while using an adaptive Lepski-type rule to select the bandwidth according to the local authorship structure. Our method is simple to implement and does not require token-level labeled data for training. Theoretically, we characterize this trade-off and show that the proposed method achieves favorable mean square error performance in estimating the underlying signal. Empirically, we demonstrate strong performance of our method against a wide range of baselines in both synthetic datasets and a realistic dataset. We deploy a publicly accessible website that implements the methods as well.