M4: Multi-Generator, Multi-Domain, and Multi-Lingual Black-Box Machine-Generated Text Detection
Large language models have led to an increase in machine-generated content, raising concerns about potential misuse. This study focuses on creating automated systems to detect machine-generated texts and address misuse.
AIpom: A Two-Stage Pipeline for Detecting Human-Machine Boundary in Mixed Text
AIpom, a novel method for human-machine mixed text detection, leverages a pipeline of decoder and encoder models to accurately identify the boundary between human-written and machine-generated text.
Detecting Machine-Generated Text: Exploring Contrastive Learning for Robust and Efficient Classification
Contrastive learning can be an effective approach for detecting machine-generated text, even with a single model and without relying on the specific text generation model used.
Detecting Machine-Generated Text Across Diverse Domains and Language Models
Combining linguistic features and language model embeddings can effectively distinguish machine-generated text from human-written text, even across unseen language models and domains.
