RAID: A Shared Benchmark for Robust Evaluation of Machine-Generated Text Detectors.
Liam DuganAlyssa HwangFilip TrhlíkAndrew ZhuJosh Magnus LudanHainiu XuDaphne IppolitoChris Callison-BurchPublished in: ACL (1) (2024)
Keyphrases
- text retrieval
- quantitative evaluation
- object detection
- database
- web documents
- human generated
- evaluation criteria
- keywords
- information retrieval
- neural network
- text mining
- language model
- computationally efficient
- semantic information
- computer vision
- partial occlusion
- robust estimation
- sentence level
- text information
- data sets