Advantages of the test are that one can assume that human judges are good representatives of human intelligence, and the test eliminates the possibility of bias withing the controlled parameters.

The goal of the Turing Test, developed by Alan Turing, is to level thinking on computer programs or artificial intelligence.Findings include: model performance and calibration both improve with scale, but are poor in absolute terms (and when compared with rater performance) performance is remarkably similar across model classes, though with benefits from sparsity tasks that improve gradually and predictably commonly involve a large knowledge or memorization component, whereas tasks that exhibit 'breakthrough' behavior at a critical scale often involve multiple steps or components, or brittle metrics social bias typically increases with scale in settings with ambiguous context, but this can be improved with prompting. In addition, a team of human expert raters performed all tasks in order to provide a strong baseline. We evaluate the behavior of OpenAI's GPT models, Google- internal dense transformer architectures, and Switch-style sparse transformers on BIG-bench, across model sizes spanning millions to hundreds of billions of parameters. BIG-bench focuses on tasks that are believed to be beyond the capabilities of current language models. Task topics are diverse, drawing problems from linguistics, childhood develop- ment, math, common-sense reasoning, biology, physics, social bias, software development, and beyond. BIG-bench currently consists of 204 tasks, contributed by 450 authors across 132 institutions. To address this challenge, we introduce the Beyond the Imitation Game benchmark (BIG- bench). In order to inform future research, prepare for disruptive new model capabilities, and ameliorate socially harmful effects, it is vital that we understand the present and near-future capabilities and limitations of language models.

Despite their potentially transformative impact, these new capabilities are as yet poorly characterized. Abstract: Language models demonstrate both quantitative improvement and new qualitative capabilities with increasing scale.

Download the imitation game test