Who Wrote This? The Key to Zero-Shot LLM-Generated Text Detection Is GECScore

arXiv:2405.04286v1 Announce Type: new
Abstract: The efficacy of an large language model (LLM) generated text detector depends substantially on the availability of sizable training data. White-box zero-shot detectors, which require no such data, are nonetheless limited by the accessibility of the source model of the LLM-generated text. In this paper, we propose an simple but effective black-box zero-shot detection approach, predicated on the observation that human-written texts typically contain more grammatical errors than LLM-generated texts. This approach entails computing the …

abstract accessibility arxiv availability box cs.cl data detection detectors generated key language language model large language large language model llm text the key training training data type zero-shot