Tracking instruction-tuned LLM openness. Paper: Liesenfeld, Andreas, Alianda Lopez, and Mark Dingemanse. 2023. “Opening up ChatGPT: Tracking Openness, Transparency, and Accountability in Instruction-Tuned Text Generators.” In Proceedings of the 5th International Conference on Conversational User Interfaces. doi:10.1145/3571884.3604316.
Twitter thread introducing this model links to most resources we'd want to check: https://twitter.com/llm360/status/1795833911580438807