Inverse scaling can become U-shaped

By Jason Wei, Yi Tay, Quoc V. Le. 2022

86 citations recorded by the source corpus.

Scaling up language models has been empirically shown to improve performance on a wide range of downstream tasks. However, if we were to observe worse performance as a function of scale ("inverse scaling") on certain tasks, this would indicate that scaling can also encourage behaviors that are misaligned with human preferences. The Inverse Scaling Prize (McKenzie et al. 2022) identified eleven such inverse scaling tasks, evaluated on models of up to 280B parameters and up to 500 zettaFLOPs of training compute. This paper takes a closer look at these inverse scaling tasks. We evaluate models of up to 540B parameters, trained on five times more compute than those evaluated in the Inverse Scaling Prize. With this increased range of model sizes and training compute, only four out of the eleven tasks remain inverse scaling. Six out of the eleven tasks exhibit"U-shaped scaling", where performance decreases up to a certain size, and then increases again up to the largest model evaluated (the one remaining task displays positive scaling). In addition, we find that 1-shot examples and chain-of-thought can help mitigate undesirable scaling patterns even further. U-shaped scaling suggests that the inverse scaling trend observed in McKenzie et al. (2022) may not continue to hold for larger models, which we attribute to the presence of distractor tasks that only sufficiently large models can avoid.

Subfields: Artificial Intelligence, Information Systems, Scaling & Learning Theory

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Information Systems

In Scaling & Learning Theory

Citations in the corpus

A Systematic Study and Comprehensive Evaluation of ChatGPT on Benchmark Datasets (2023, methodology): “Performance in Inverse Scaling Tasks: For inverse scaling (Wei et al., 2022b), we evaluate the performance of two versions of ChatGPT: (i) the December 15 version in chat.openai.com and (ii) the latest API version gpt-3.”
Harnessing the Power of LLMs in Practice: A Survey on ChatGPT and Beyond (2023, background): “From this perspective, the U-shape phenomenon can be seen as a transformation of the inverse-scaling phenomenon due to some emergent abilities from sufficiently large models [124].”
On Second Thought, Let’s Not Think Step by Step! Bias and Toxicity in Zero-Shot Reasoning (2022, background): “We are un-sure if our documented effect is U-shaped (Wei et al., 2022a)—specifically, if further increasing 2022)).”
Inverse Scaling: When Bigger Isn't Better (2023, methodology): “The Inverse Scaling Prize tasks helped drive the discovery of U-shaped scaling trends (Wei et al., 2022a), where inverse scaling trends reversed at sufficient model scale.”