ChatGPT

Overviews

Evaluations on Benchmarks
- Bang et al 2023 - A Multitask, Multilingual, Multimodal Evaluation of ChatGPT on Reasoning, Hallucination, and Interactivity
- Laskar et al 2023 - A Systematic Study and Comprehensive Evaluation of ChatGPT on Benchmark Datasets (ACL 2023)
Impact of ChatGPT
- Estimating Prevalence of Usage
  - Kobak et al 2024 - Delving into ChatGPT usage in academic writing through excess vocabulary (Talks about the word delve)
Datasets
- (Interaction logs, etc)
- Zhao et al 2024 - WildChat: 1M ChatGPT Interaction Logs in the Wild dataset