GooAQ: Open Question Answering with Diverse Answer Types
Daniel Khashabi, Amos H.C. Ng, Tushar Khot, Ashish Sabharwal, Hannaneh Hajishirzi, Chris Callison-Burch · 2021
While day-to-day questions come with a variety of answer types, the current open questionanswering (QA) literature represents isolated efforts on niche response types, with a heavy focus on specific kinds of short responses (people, places, etc.).To address this gap, we present GOOAQ, a large-scale dataset collected from Google questions and answers, containing 3 million questions with diverse answer types ranging from factual short answers to snippets to collections.Our human evaluation shows that 94% of the mined answers are accurate, enabling fine-tuning a pre-trained language model for answering GOOAQ questions.We use this dataset to study inherent differences between models producing different answer types, and observe interesting trends.For example, in line with recent work, LM's strong performance on GOOAQ's shortanswer questions heavily benefits from annotated data.However, their surprisingly high quality in generating coherent and accurate answers for questions requiring long responses (such as 'how' and 'why' questions) is less reliant on observing annotated data and mainly supported by their pre-training.Moreover, we show that GOOAQ is a valuable training resource, resulting in strong performance on the recent ELI5 long-answers dataset.We release GOOAQ to facilitate further research on improving QA with diverse response types. 1Question: how many calories burned 30 minutes crossfit?Answer(short: from the snippet): 260 calories Answer(snippet): According to the American Council on Exercise, a 115-pound person running for 30 minutes at a slow-to-moderate pace (a 10-minute mile) would burn about 260 calories. Question: what are the steps for decision making? Answer(collection): [Step 1: Identify the decision You realize that you need to make a decision. , Step 2: Gather relevant information. Step 3: Identify the alternatives. Step 4: Weigh the evidence. Step 5: Choose among alternatives. Step 6: Take action. Step 7: Review your decision & its consequences.