Built independently by an author, for readers. Read the story and support ChapterPal
ChapterPal Logo
Log inSign up

keyword

word-class distribution

Word-class distribution refers to the statistical frequency, proportion, and relative occurrence of different grammatical categories, such as nouns, verbs, adjectives, and conjunctions, within a specific text or corpus. In natural language processing, corpus linguistics, and stylometry, this distribution characterizes the syntactic structure and grammatical composition of written or spoken language. Because individual authors, registers, and text generation systems often exhibit distinct habits in how heavily they rely on specific parts of speech, measuring the distribution of word classes serves as an essential method for evaluating syntactic complexity, analyzing writing style, and performing tasks such as genre identification, authorship attribution, and text classification.

1 item

Token Prediction as Implicit Classification to Identify LLM-Generated Text

Token Prediction as Implicit Classification to Identify LLM-Generated Text

Yutian Chen, Hao Kang, Vivian Zhai, Liangze Li, Rita Singh, Bhiksha Raj

OrganizationsCarnegie Mellon University

Why you should read this

Presents a method that reframes machine-generated text attribution as next-token prediction rather than explicit classification, outperforming traditional classifier heads and capturing model-specific writing styles across a 340,000-sample dataset.

This paper introduces a novel approach for identifying the possible large language models (LLMs) involved in text generation. Instead of adding an additional classification layer to a base LM, we reframe the classification task as a next-token prediction task and directly fine-tune the base LM to perform it. We utilize the Text-to-Text Transfer Transformer (T5) model as the backbone for our experiments. We compared our approach to the more direct approach of utilizing hidden states for classification. Evaluation shows the exceptional performance of our method in the text classification task, highlighting its simplicity and efficiency. Furthermore, interpretability studies on the features extracted by our model reveal its ability to differentiate distinctive writing styles among various LLMs even in the absence of an explicit classifier. We also collected a dataset named OpenLLMText, containing approximately 340k text samples from human and LLMs, including GPT3.5, PaLM, LLaMA, and GPT2.

Added

2026-10-03