Beyond the Imitation Game: Quantifying and extrapolating the capabilities of language models
Beyond the Imitation Game: Quantifying and extrapolating the capabilities of language models
复制标题
DOI：
--
复制
发表时间：
2022-06
期刊：
影响因子：
--
通讯作者：
Aarohi Srivastava;Abhinav Rastogi;Abhishek Rao;Abu Awal Md Shoeb;Abubakar Abid;Adam Fisch;Adam R. Brown-Ad
中科院分区：
文献类型：
--
作者：
Aarohi Srivastava;Abhinav Rastogi;Abhishek Rao;Abu Awal Md Shoeb;Abubakar Abid;Adam Fisch;Adam R. Brown-Ad
随着规模的增加，语言模型表现出定量的改进和新的定性能力。尽管这些新的能力具有潜在的变革性影响，但其特点尚不明确。为了为未来的研究提供信息，为破坏性的新模型能力做好准备，并减轻社会有害影响，我们必须了解语言模型的当前和不久的将来的能力和局限性。为了应对这一挑战，我们引入了超越模仿游戏基准（BIG-bench）。BIG-bench目前包含204个任务，由132个机构的450位作者贡献。任务主题是多样的，从语言学，儿童发展，数学，常识推理，生物学，物理学，社会偏见，软件开发等方面提出问题。BIG-bench专注于被认为超出当前语言模型能力的任务。我们评估了OpenAI的GPT模型、Google内部的密集Transformer架构以及BIG-bench上的Switch风格稀疏transformer的行为，这些模型大小跨越数百万到数千亿个参数。此外，一个由人类专家评分员组成的团队执行了所有任务，以提供一个强大的基线。调查结果包括：模型性能和校准都随着规模的增加而提高，但绝对值很差（与评分员的表现相比）;不同模型类别的表现非常相似，尽管稀疏性带来了好处;逐步改进和可预测的任务通常涉及大量的知识或记忆成分，而在关键尺度上表现出“突破“行为的任务通常涉及多个步骤或成分，或脆弱的指标;社会偏见通常会在具有模糊背景的环境中随着规模的增加而增加，但这可以通过提示来改善。
Language models demonstrate both quantitative improvement and new qualitative capabilities with increasing scale. Despite their potentially transformative impact, these new capabilities are as yet poorly characterized. 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. To address this challenge, we introduce the Beyond the Imitation Game benchmark (BIG-bench). BIG-bench currently consists of 204 tasks, contributed by 450 authors across 132 institutions. Task topics are diverse, drawing problems from linguistics, childhood development, math, common-sense reasoning, biology, physics, social bias, software development, and beyond. BIG-bench focuses on tasks that are believed to be beyond the capabilities of current language models. 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. In addition, a team of human expert raters performed all tasks in order to provide a strong baseline. 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.