Ask Me Anything: A simple strategy for prompting language models

By Simran Arora, Avanika Narayan, Mayee F. Chen, Laurel J. Orr, Neel Guha, Kush S. Bhatia, Ines Chami, Frederic Sala, Christopher R'e. 2022

275 citations recorded by the source corpus.

Large language models (LLMs) transfer well to new tasks out-of-the-box simply given a natural language prompt that demonstrates how to perform the task and no additional training. Prompting is a brittle process wherein small modifications to the prompt can cause large variations in the model predictions, and therefore significant effort is dedicated towards designing a painstakingly"perfect prompt"for a task. To mitigate the high degree of effort involved in prompt-design, we instead ask whether producing multiple effective, yet imperfect, prompts and aggregating them can lead to a high quality prompting strategy. Our observations motivate our proposed prompting method, ASK ME ANYTHING (AMA). We first develop an understanding of the effective prompt formats, finding that question-answering (QA) prompts, which encourage open-ended generation ("Who went to the park?") tend to outperform those that restrict the model outputs ("John went to the park. Output True or False."). Our approach recursively uses the LLM itself to transform task inputs to the effective QA format. We apply the collected prompts to obtain several noisy votes for the input's true label. We find that the prompts can have very different accuracies and complex dependencies and thus propose to use weak supervision, a procedure for combining the noisy predictions, to produce the final predictions for the inputs. We evaluate AMA across open-source model families (e.g., EleutherAI, BLOOM, OPT, and T0) and model sizes (125M-175B parameters), demonstrating an average performance lift of 10.2% over the few-shot baseline. This simple strategy enables the open-source GPT-J-6B model to match and exceed the performance of few-shot GPT3-175B on 15 of 20 popular benchmarks. Averaged across these tasks, the GPT-J-6B model outperforms few-shot GPT3-175B. We release our code here: https://github.com/HazyResearch/ama_prompting

Subfields: Artificial Intelligence, Health Informatics, Large Language Models, Prompting & Reasoning

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 Health Informatics

In Large Language Models

In Prompting & Reasoning

Citations in the corpus

FrugalGPT: How to Use Large Language Models While Reducing Cost and Improving Performance (2023, unknown): “Directing 86% of queries to GPT-J and the remaining 14% to GPT-4 is considerably more cost-effective and performant than relying solely on GPT-4.”
Distilling Step-by-Step! Outperforming Larger Language Models with Less Training Data and Smaller Model Sizes (2023, methodology): “It supports learning from limited labeled data, since the larger teacher model is often used to generate a training dataset with noisy pseudo labels (Chen et al., 2020; Il-iopoulos et al., 2022; Wang et al., 2021; Smith et al., 2022a; Arora et al., 2022; Agrawal et al., 2022).”
Can Foundation Models Wrangle Your Data? (2022, background): “Excitingly, recent work [14] has demonstrated that techniques such as prompt ensembling and prompt reframing can enable open-source models like GPT-J-6B [95] to out-perform GPT3-175B on popular natural language understanding benchmarks.”
Hyena Hierarchy: Towards Larger Convolutional Language Models (2023, background): “, 2019) with the parsing pipeline of (Arora et al., 2022).”
Large Language Models are Better Reasoners with Self-Verification (2022, background): “It has been extensively studied in order to evaluate the various reasoning abilities of language models (Arora et al., 2022; Madaan et al., 2022), including arithmetic reasoning (Koncel-Kedziorski et al.”
A Prompt Pattern Catalog to Enhance Prompt Engineering with ChatGPT (2023, background): “Due to the convincing nature of this generated text, however, users may not perform appropriate due diligence to determine its accuracy.”