In this task, participants are provided with a text generated automatically and must determine which model generated it. To encourage models to learn features that generalize to new writing styles, five domains are used for training and two different domains for testing. The texts have been generated by LLMs trained by various providers, such as OpenAI, Amazon Bedrock, Anthropic, Cohere, AI21, Google Vertex AI, Meta, among others.
Publication
Sarvazyan et al. (2024). Overview of IberAuTexTification at IberLEF 2024: Detection and Attribution of Machine-Generated Text on Languages of the Iberian Peninsula. Procesamiento del Lenguaje Natural, 73: 421-434.
Language
Spanish
NLP topic
Dataset
Year
2024
Publication link
Task results
| System | MacroF1 Sort ascending |
|---|---|
| gmc_fosunlp_1 | 0.5231 |
| iimasNLP_3 | 0.5173 |
| Drocks_2 | 0.5075 |
| Drocks_1 | 0.5030 |
| iimasNLP_2 | 0.4958 |
