An official website of the United States government
Long-form answers, consisting of multiple sentences, can provide nuanced and comprehensive answers to a broader set of questions. To better understand this complex and understudied task, we study the functional structure of long-form answers collected from three datasets, ELI5, WebGPT and Natural Questions. Our main goal is to understand how humans organize information to craft complex answers. We develop an ontology of six sentence-level functional roles for long-form answers, and annotate 3.9k sentences in 640 answer paragraphs. Different answer collection methods manifest in different discourse structures. We further analyze model-generated answers – finding that annotators agree less with each other when annotating model-generated answers compared to annotating human-written answers. Our annotated data enables training a strong classifier that can be used for automatic analysis. We hope our work can inspire future research on discourse-level modeling and evaluation of long-form QA systems.
more »
« less
GooAQ: Open Question Answering with Diverse Answer Types
While day-to-day questions come with a variety of answer types, the current question-answering (QA) literature has failed to adequately address the answer diversity of questions. To this end, we present GooAQ, a large-scale dataset with a variety of answer types. This dataset contains over 5 million questions and 3 million answers collected from Google. GooAQ questions are collected semi-automatically from the Google search engine using its autocomplete feature. This results in naturalistic questions of practical interest that are nonetheless short and expressed using simple language. GooAQ answers are mined from Google’s responses to our collected questions, specifically from the answer boxes in the search results. This yields a rich space of answer types, containing both textual answers (short and long) as well as more structured ones such as collections. We benchmark T5 models on GooAQ and observe that: (a) in line with recent work, LM’s strong performance on GooAQ’s short-answer questions heavily benefit from annotated data; however, (b) their quality in generating coherent and accurate responses 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. We release GooAQ to facilitate further research on improving QA with diverse response types.
more »
« less
- Award ID(s):
- 1928474
- PAR ID:
- 10344228
- Date Published:
- Journal Name:
- Findings of the Association for Computational Linguistics: EMNLP 2021
- Page Range / eLocation ID:
- 421 to 433
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Naturally-occurring information-seeking questions often contain questionable assumptions -- assumptions that are false or unverifiable. Questions containing questionable assumptions are challenging because they require a distinct answer strategy that deviates from typical answers to information-seeking questions. For instance, the question "When did Marie Curie discover Uranium?" cannot be answered as a typical when question without addressing the false assumption "Marie Curie discovered Uranium". In this work, we propose (QA)2 (Question Answering with Questionable Assumptions), an open-domain evaluation dataset consisting of naturally-occurring search engine queries that may or may not contain questionable assumptions. To be successful on (QA)2, systems must be able to detect questionable assumptions and also be able to produce adequate responses for both typical information-seeking questions and ones with questionable assumptions. We find that current models do struggle with handling questionable assumptions -- the best performing model achieves 59% human rater acceptability on abstractive QA with (QA)2 questions, leaving substantial headroom for progress.more » « less
-
Human actions or lack thereof contribute to a large majority of cybersecurity incidents. Traditionally, when looking for advice on cybersecurity questions, people have turned to search engines or social sites like Reddit. The rapid adoption of chatbot technologies is offering a potentially more direct way of getting similar advice. Initial research suggests, however, that while chatbot answers to common cybersecurity questions tend to be fairly accurate, they may not be very effective as they often fall short on other desired qualities such as understandability, actionability, or motivational power. Research in this area thus far has been limited to the evaluation by researchers themselves on a small number of synthetic questions. This article reports on what we believe to be the first in situ evaluation of a cybersecurity Question Answering (QA) assistant. We also evaluate a prompt engineered to help the cybersecurity QA assistant generate more effective answers. The study involved a 10-day deployment of a cybersecurity QA assistant in the form of a Chrome extension. Collectively, participants (N=51) evaluated answers generated by the assistant to over 1,000 cybersecurity questions they submitted as part of their regular day-to-day activities. The results suggest that a majority of participants found the assistant useful and often took actions based on the answers they received. In particular, the study indicates that prompting successfully improved the effectiveness of answers and, in particular, the likelihood that users follow their recommendations (fraction of participants who actually followed the advice was 0.514 with prompting vs. 0.402 without prompting, p=4.61E-04), an impact on people’s actual behavior. We provide a detailed analysis of data collected in this study, discuss their implications, and outline next steps in the development and deployment of effective cybersecurity QA assistants that offer the promise of changing actual user behavior and of reducing human-related security incidents.more » « less
-
Human actions or lack thereof contribute to a large majority of cybersecurity incidents. Traditionally, when looking for advice on cybersecurity questions, people have turned to search engines or social sites like Reddit. The rapid adoption of chatbot technologies is offering a potentially more direct way of getting similar advice. Initial research suggests, however, that while chatbot answers to common cybersecurity questions tend to be fairly accurate, they may not be very effective as they often fall short on other desired qualities such as understandability, actionability, or motivational power. Research in this area thus far has been limited to the evaluation by researchers themselves on a small number of synthetic questions. This article reports on what we believe to be the first in situ evaluation of a cybersecurity Question Answering (QA) assistant. We also evaluate a prompt engineered to help the cybersecurity QA assistant generate more effective answers. The study involved a 10-day deployment of a cybersecurity QA assistant in the form of a Chrome extension. Collectively, participants (N=51) evaluated answers generated by the assistant to over 1,000 cybersecurity questions they submitted as part of their regular day-to-day activities. The results suggest that a majority of participants found the assistant useful and often took actions based on the answers they received. In particular, the study indicates that prompting successfully improved the effectiveness of answers and, in particular, the likelihood that users follow their recommendations (fraction ofparticipants who actually followed the advice was 0.514 with prompting vs. 0.402 without prompting, p=4.61E-04), an impacton people’s actual behavior. We provide a detailed analysis of data collected in this study, discuss their implications, and outline next steps in the development and deployment of effective cybersecurity QA assistants that offer the promise of changing actual user behavior and of reducing human-related security incidents.more » « less
-
How do Google Search results change following an impactful real-world event, such as the U.S. Supreme Court decision on June 24, 2022 to overturn Roe v. Wade? And what do they tell us about the nature of event-driven content, generated by various participants in the online information environment? In this paper, we present a dataset of more than 1.74 million Google Search results pages collected between June 24 and July 17, 2022, intended to capture what Google Search surfaced in response to queries about this event of national importance. These search pages were collected for 65 locations in 13 U.S. states, a mix of red, blue, and purple states, with respect to their voting patterns. We describe the process of building a set of circa 1,700 phrases used for searching Google, how we gathered the search results for each location, and how these results were parsed to extract information about the most frequently encountered web domains. We believe that this dataset, which comprises raw data (search results as HTML files) and processed data (extracted links organized as CSV files) can be used to answer research questions that are of interest to computational social scientists as well as communication and media studies scholars.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Conference Paper:
- https://doi.org/10.18653/v1/2021.findings-emnlp.38
