STREET: A Multi-Task Structured Reasoning and Explanation Benchmark
This provides a new benchmark for training and testing multi-step reasoning and explanation systems in natural language, addressing a gap in existing QA datasets.
The authors tackled the problem of evaluating natural language reasoning and explanation by introducing STREET, a benchmark requiring models to provide step-by-step structured explanations alongside answers, and found that models like GPT-3 and T5 lag behind human performance.
We introduce STREET, a unified multi-task and multi-domain natural language reasoning and explanation benchmark. Unlike most existing question-answering (QA) datasets, we expect models to not only answer questions, but also produce step-by-step structured explanations describing how premises in the question are used to produce intermediate conclusions that can prove the correctness of a certain answer. We perform extensive evaluation with popular language models such as few-shot prompting GPT-3 and fine-tuned T5. We find that these models still lag behind human performance when producing such structured reasoning steps. We believe this work will provide a way for the community to better train and test systems on multi-step reasoning and explanations in natural language.