Researchers have developed a new set of models for detecting AI-generated text, focusing on fine-grained analysis and human-LLM co-authored content. These models are trained on a large dataset of over 2.4 million texts, primarily co-authored by proprietary LLMs across 23 languages. The system demonstrates robust performance on unseen domains, generators, non-native speakers, and adversarial inputs, offering detailed comparisons against various adversarial methods and input characteristics. AI
IMPACT Enhances the ability to identify AI-generated content, crucial for maintaining authenticity and trust in digital communication.
RANK_REASON The cluster contains a research paper published on arXiv detailing new models and a dataset for detecting AI-generated text. [lever_c_demoted from research: ic=1 ai=1.0]
- alphaXiv
- arXiv
- arXivLabs
- CatalyzeX Code Finder for Papers
- DagsHub
- Gotit.pub
- Hugging Face
- human-LLM co-authored texts
- Proprietary LLMs
- Ram Mohan Rao Kadiyala
- ScienceCast
- token classification
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →
