Jump to Content

Xuezhi Wang

Xuezhi Wang is a Research Scientist at Google DeepMind. Her primary interests are reasoning and systematic generalization in language models. She has also worked on robustness and fairness in NLP models. Xuezhi received her PhD degree from the Computer Science Department in Carnegie Mellon University in 2016.
Authored Publications
Google Publications
Other Publications
Sort By
  • Title
  • Title, descending
  • Year
  • Year, descending
    A Mixed-Methods Approach to Understanding User Trust after Voice Assistant Failures
    Allison Mercurio
    Amanda Elizabeth Baughan
    Jilin Chen
    Proceedings of the 2023 CHI Conference on Human Factors in Computing Systems Pages (2023)
    Preview abstract Despite huge gains in performance in natural language understanding via large language models in recent years, voice assistants still often fail to meet user expectations. In this study, we conducted a mixed-methods analysis of how voice assistant failures affect users' trust in their voice assistants. To illustrate how users have experienced these failures, we contribute a crowdsourced dataset of 199 voice assistant failures, categorized across 12 failure sources. Relying on interview and survey data, we find that certain failures, such as those due to overcapturing users' input, derail user trust more than others. We additionally examine how failures impact users' willingness to rely on voice assistants for future tasks. Users often stop using their voice assistants for specific tasks that result in failures for a short period of time before resuming similar usage. We demonstrate the importance of low stakes tasks, such as playing music, towards building trust after failures. View details
    Preview abstract Existing pre-trained models are generally geared towards a particular class of problems. To date, there seems to be still no consensus on what the right architecture and pre-training setup should be. This paper presents a unified framework for pre-training models that are universally effective across datasets and setups. We begin by disentangling architectural archetypes with pre-training objectives – two concepts that are commonly conflated. Next, we present a generalized and unified perspective for self-supervision in NLP and show how different pre-training objectives can be cast as one another and how interpolating between different objectives can be effective. We then propose Mixture-of-Denoisers (MoD), a pretraining objective that combines diverse pre-training paradigms together. We furthermore introduce a notion of mode switching, wherein downstream fine-tuning is associated with specific pre-training schemes. We conduct extensive ablative experiments to compare multiple pre-training objectives and find that our method pushes the Pareto-frontier by outperforming T5 and/or GPT-like models across multiple diverse setups. Finally, by scaling our model up to 20B parameters, we achieve SOTA performance on 50 well-established supervised NLP tasks ranging from language generation (with automated and human evaluation), language understanding, text classification, question answering, commonsense reasoning, long text reasoning, structured knowledge grounding and information retrieval. Our model also achieve strong results at in-context learning, outperforming 175B GPT-3 on zero-shot SuperGLUE and tripling the performance of T5-XXL on oneshot summarization. Finally, we show that UL2 20B works well with chain-ofthought prompting and reasoning tasks, making it an appealing choice for research into reasoning at a small to medium scale of 20B parameters. We publicly release Flax-based T5X model checkpoints for the 20B model. View details
    Recitation-Augmented Language Models
    Zhiqing Sun
    Yi Tay
    Yiming Yang
    ICLR 2023 (to appear)
    Preview abstract We propose a new paradigm to help Large Language Models (LLMs) generate more accurate factual knowledge without retrieving from an external corpus, called RECITation-augmented gEneration (RECITE). Different from retrieval-augmented language models that retrieve relevant documents before generating the outputs, given an input, RECITE first recites one or several relevant passages from LLMs' own memory via sampling, and then produces the final answers. We show that RECITE is a powerful paradigm for knowledge-intensive NLP tasks. Specifically, we show that by utilizing recitation as the intermediate step, a recite-and-answer scheme can achieve new state-of-the-art performance in various closed-book question answering (CBQA) tasks. In experiments, we verify the effectiveness of RECITE on three pre-trained models (PaLM, UL2, and OPT) and three CBQA tasks (Natural Questions, TriviaQA, and HotpotQA). View details
    Preview abstract Chain-of-thought prompting combined with pre-trained large language models has achieved encouraging results on complex reasoning tasks. In this paper, we propose a new decoding strategy, self-consistency, to replace the naive greedy decoding used in chain-of-thought prompting. It first samples a diverse set of reasoning paths instead of only taking the greedy one, and then selects the most consistent answer by marginalizing out the sampled reasoning paths. Self-consistency leverages the intuition that a complex reasoning problem typically admits multiple different ways of thinking leading to its unique correct answer. Our extensive empirical evaluation shows that self-consistency boosts the performance of chain-of-thought prompting with a striking margin on a range of popular arithmetic and commonsense reasoning benchmarks, including GSM8K (+17.9%), SVAMP (+11.0%), AQuA (+12.2%), StrategyQA (+6.4%) and ARC-challenge (+3.9%). View details
    TEMPERA: Test-Time Prompt Editing via Reinforcement Learning
    Tianjun Zhang
    Dale Schuurmans
    Joseph E. Gonzalez
    ICLR 2023 (to appear)
    Preview abstract Careful prompt design is critical to the use of large language models in zero-shot or few-shot learning. As a consequence, there is a growing interest in automated methods to design optimal prompts. In this work, we propose Test-time Prompt Editing using Reinforcement learning (TEMPERA). In contrast to prior prompt generation methods, TEMPERA can efficiently leverage prior knowledge, is adaptive to different queries and provides an interpretable prompt for every query. To achieve this, we design a novel action space that allows flexible editing of the initial prompts covering a wide set of commonly-used components like instructions, few-shot exemplars, and verbalizers. The proposed method achieves significant gains compared with recent SoTA approaches like prompt tuning, AutoPrompt, and RLPrompt, across a variety of tasks including sentiment analysis, topic classification, natural language inference, and reading comprehension. Our method achieves 5.33x on average improvement in sample efficiency when compared to the traditional fine-tuning methods. View details
    Preview abstract Recently, NLP models have achieved remarkable progress across a variety of tasks; however, they have also been criticized for being not robust. Many robustness problems can be attributed to models exploiting "spurious correlations", or "shortcuts" between the training data and the task labels. Most existing work identifies a limited set of task-specific shortcuts via human priors or error analyses, which requires extensive expertise and efforts. In this paper, we aim to automatically identify such spurious correlations in NLP models at scale. We first leverage existing interpretability methods to extract tokens that significantly affect model's decision process from the input text. We then distinguish "genuine" tokens and "spurious" tokens by analyzing model predictions across multiple corpora and further verify them through knowledge-aware perturbations. We show that our proposed method can effectively and efficiently identify a scalable set of ``shortcuts'', and mitigating these leads to more robust models in multiple applications. View details
    Preview abstract Large pre-trained language models have shown remarkable performance over the past few years. These models, however, sometimes learn superficial features from the dataset and cannot generalize to the distributions that are dissimilar to the training scenario. There have been several approaches proposed to reduce model's reliance on these bias features which can improve model robustness in the out-of-distribution setting. However, existing methods usually use a fixed low-capacity model to deal with various bias features, which ignore the learnability of those features. In this paper, we analyze a set of existing bias features and demonstrate there is no single model that works best for all the cases. We further show that by choosing an appropriate bias model, we can obtain a better robustness result than baselines with a more sophisticated model design. View details
    PaLM: Scaling Language Modeling with Pathways
    Aakanksha Chowdhery
    Sharan Narang
    Jacob Devlin
    Maarten Bosma
    Hyung Won Chung
    Sebastian Gehrmann
    Parker Schuh
    Sasha Tsvyashchenko
    Abhishek Rao
    Yi Tay
    Noam Shazeer
    Nan Du
    Reiner Pope
    James Bradbury
    Guy Gur-Ari
    Toju Duke
    Henryk Michalewski
    Xavier Garcia
    Liam Fedus
    David Luan
    Barret Zoph
    Ryan Sepassi
    David Dohan
    Shivani Agrawal
    Mark Omernick
    Marie Pellat
    Aitor Lewkowycz
    Erica Moreira
    Rewon Child
    Oleksandr Polozov
    Zongwei Zhou
    Brennan Saeta
    Michele Catasta
    Jason Wei
    Slav Petrov
    arxiv:2204.02311 (2022)
    Preview abstract Large language models have been shown to achieve remarkable performance across a variety of natural language tasks using few-shot learning, which drastically reduces the number of task-specific training examples needed to adapt the model to a particular application. To further our understanding of the impact of scale on few-shot learning, we trained a 540-billion parameter, densely activated, Transformer language model, which we call Pathways Language Model PaLM. We trained PaLM on 6144 TPU v4 chips using Pathways, a new ML system which enables highly efficient training across multiple TPU Pods. We demonstrate continued benefits of scaling by achieving state-of-the-art few-shot learning results on hundreds of language understanding and generation benchmarks. On a number of these tasks, PaLM 540B achieves breakthrough performance, outperforming the finetuned state-of-the-art on a suite of multi-step reasoning tasks, and outperforming average human performance on the recently released BIG-bench benchmark. A significant number of BIG-bench tasks showed discontinuous improvements from model scale, meaning that performance steeply increased as we scaled to our largest model. PaLM also has strong capabilities in multilingual tasks and source code generation, which we demonstrate on a wide array of benchmarks. We additionally provide a comprehensive analysis on bias and toxicity, and study the extent of training data memorization with respect to model scale. Finally, we discuss the ethical considerations related to large language models and discuss potential mitigation strategies. View details
    Preview abstract Continual learning is essential for real-world deployment when there is a need to quickly adapt the model to new tasks without forgetting knowledge of old tasks. Existing work on continual sequence generation either always reuses existing parameters to learn new tasks, which is vulnerable to catastrophic forgetting on dissimilar tasks, or blindly adds new parameters for every new task, which could prevent knowledge sharing between similar tasks. To get the best of both worlds, in this work, we propose continual sequence generation with adaptive compositional modules to adaptively add modules in transformer architectures and compose both old and new modules for new tasks. We also incorporate pseudo experience replay to facilitate knowledge transfer in those shared modules. Experiment results on various sequences of generation tasks show that our framework can adaptively add modules or reuse modules based on task similarity, outperforming state-of-the-art baselines in terms of both performance and parameter efficiency. We make our code public at https://github.com/GT-SALT/Adaptive-Compositional-Modules. View details
    Preview abstract As NLP models achieved state-of-the-art performances over benchmarks and gained wide applications, it has been increasingly important to ensure the safe deployment of these models in the real world, e.g., making sure the models are robust against unseen or challenging scenarios. Despite robustness being an increasingly studied topic, it has been separately explored in applications like vision and NLP, with various definitions, evaluation and mitigation strategies in multiple lines of research. In this paper, we aim to provide a unifying survey of how to define, measure and improve robustness in NLP. We first connect multiple definitions of robustness, then unify various lines of work on identifying robustness failures and evaluating models' robustness. Correspondingly, we present mitigation strategies that are data-driven, model-driven, and inductive-prior-based, with a more systematic view of how to effectively improve robustness in NLP models. Finally, we conclude by outlining open challenges and future directions to motivate further research in this area. View details
    Preview abstract Developing robust NLP models that perform well on many, even small, slices of data is a significant but important challenge, with implications from fairness to general reliability. To this end, recent research has explored how models rely on spurious correlations, and how counterfactual data augmentation (CDA) can mitigate such issues. In this paper we study how and why modeling counterfactuals over multiple attributes can go significantly further in improving model performance. We propose RDI, a context-aware methodology which takes into account the impact of secondary attributes on the model’s predictions and increases sensitivity for secondary attributes over reweighted counterfactually augmented data. By implementing RDI in the context of toxicity detection, we find that accounting for secondary attributes can significantly improve robustness, with improvements in sliced accuracy on the original dataset up to 7% compared to existing robustness methods. We also demonstrate that RDI generalizes to the coreference resolution task and provide guidelines to extend this to other tasks. View details
    Preview abstract Continual learning has become increasingly important as it enables NLP models to constantly learn and gain knowledge over time. Previous continual learning methods are mainly designed to preserve knowledge from previous tasks, without much emphasis on how to well generalize models to new tasks. In this work, we propose an information disentanglement based regularization method for continual learning on text classification. Our proposed method first disentangles text hidden spaces into representations that are generic to all tasks and representations specific to each individual task, and further regularizes these representations differently to better constrain the knowledge required to generalize. We also introduce two simple auxiliary tasks: next sentence prediction and task-id prediction, for learning better generic and specific representation spaces. Experiments conducted on large-scale benchmarks demonstrate the effectiveness of our method in continual text classification tasks with various sequences and lengths over state-of-the-art baselines. We have publicly released our code at https://github.com/GT-SALT/IDBR. View details
    Preview abstract Most literature in fairness has focused on improving fairness with respect to one single model or one single objective. However, real-world machine learning systems are usually composed of many different components. Unfortunately, recent research has shown that even if each component is "fair", the overall system can still be "unfair". In this paper, we focus on how well fairness composes over multiple components in real systems. We consider two recently proposed fairness metrics for rankings: exposure and pairwise ranking accuracy gap. We provide theory that demonstrates a set of conditions under which fairness of individual models does compose. We then present an analytical framework for both understanding whether a system's signals can achieve compositional fairness, and diagnosing which of these signals lowers the overall system's end-to-end fairness the most. Despite previously bleak theoretical results, on multiple data-sets -- including a large-scale real-world recommender system -- we find that the overall system's end-to-end fairness is largely achievable by improving fairness in individual components. View details
    Preview abstract Neural networks lack adversarial robustness, ie, they are vulnerable to adversarial examples that through small perturbations to inputs cause incorrect predictions. Further, trust is undermined when models give miscalibrated predictions, ie, the predicted probability is not a good indicator of how much we should trust our model. In this paper, we study the connection between adversarial robustness and calibration and find that the inputs for which the model is sensitive to small perturbations (are easily attacked) are more likely to have poorly calibrated predictions. Based on this insight, we examine if calibration can be improved by addressing those adversarially unrobust inputs. To this end, we propose Adversarial Robustness based Adaptive Label Smoothing (AR-AdaLS) that integrates the correlations of adversarial robustness and calibration into training by adaptively softening labels for an example based on how easily it can be attacked by an adversary. We find that our method, taking the adversarial robustness of the in-distribution data into consideration, leads to better calibration over the model even under distributional shifts. In addition, AR-AdaLS can also be applied to an ensemble model to further improve model calibration. View details
    Preview abstract As multi-task models gain popularity in a wider range of machine learning applications, it is becoming increasingly important for practitioners to understand the fairness implications associated with those models. Most existing fairness literature focuses on learning a single task more fairly, while how ML fairness interacts with multiple tasks in the joint learning setting is largely under-explored. In this paper, we are concerned with how group fairness (e.g., equal opportunity, equalized odds) as an ML fairness concept plays out in the multi-task scenario. In multi-task learning, several tasks are learned jointly to exploit task correlations for a more efficient inductive transfer. This presents a multi-dimensional Pareto frontier on (1) the trade-off between group fairness and accuracy with respect to each task, as well as (2) the trade-offs across multiple tasks. We aim to provide a deeper understanding on how group fairness interacts with accuracy in multi-task learning, and we show that traditional approaches that mainly focus on optimizing the Pareto frontier of multi-task accuracy might not perform well on fairness goals. We propose a new set of metrics to better capture the multi-dimensional Pareto frontier of fairness-accuracy trade-offs uniquely presented in a multi-task learning setting. We further propose a Multi-Task-Aware Fairness (MTA-F) approach to improve fairness in multi-task learning. Experiments on several real-world datasets demonstrate the effectiveness of our proposed approach. View details
    Measuring Model Fairness under Noisy Covariates: A Theoretical Perspective
    Aditee Ajit Kumthekar
    Alex Beutel
    Jilin Chen
    Li Wei
    Nick Blumm
    Pranjal Awasthi
    Trevor Potter
    AIES (2021)
    Preview abstract In this work we study the problem of measuring the fairness of a machine learning model under noisy information. In many applications, evaluating a model according to a well-specified metric such as the FPR requires access to variables that cannot be jointly observed in a given practical setting. A standard workaround is to then use proxies for one or more of these variables. These proxies are either obtained using domain expertise or by training another machine learning model. Prior works have demonstrated the dangers of using such an approach, and strong independence assumptions are needed to provide guarantees on the accuracy of the noisy estimates via proxies. In contrast, in this work we present a general theoretical framework that aims to characterize weaker conditions under which accurate model auditing is possible via the above approach. Furthermore, our theory identifies potential sources of errors and decouples them into two interpretable parts Epsilon_c and Epsilon_g. The first part depends on natural properties of the proxy such as precision and recall, whereas the second part captures correlations between different variables of interest. We show that in many scenarios the error in the estimates is dominated by the Epsilon_c via a linear dependence, whereas the dependence on the correlations only constitutes a lower order term. As a result we expand the understanding of scenarios where model auditing via proxies can be an effective approach. Finally, we compare via simulations the theoretical upper-bounds to the distribution of simulated estimation errors and show that both theoretical guarantees and empirical results significantly improve as we progressively enforce structure along the conditions highlighted by the theory. View details
    Evaluating Fairness of Machine Learning Models Under Uncertain and Incomplete Information
    Pranjal Awasthi
    Alex Beutel
    Matthaeus Kleindessner
    Jamie Morgenstern
    Proceedings of the 2021 ACM Conference on Fairness, Accountability, and Transparency
    Preview abstract Training and evaluation of fair classifiers is a challenging problem. This is partly due to the fact that most fairness metrics of interest depend on both the sensitive attribute information and label information of the data points. In many scenarios it is not possible to collect large datasets with such information. An alternate approach that is commonly used is to separately train an attribute classifier on data with sensitive attribute information, and then use it later in the ML pipeline to evaluate the bias of a given classifier. While such decoupling helps alleviate the problem of demographic scarcity, it raises several natural questions such as: how should the attribute classifier be trained?, and how should one use a given attribute classifier for accurate bias estimation? In this work we study this question from both theoretical and empirical perspectives. View details
    ToTTo: A Controlled Table-To-Text Generation Dataset
    Sebastian Gehrmann
    Bhuwan Dhingra
    Manaal Faruqui
    Diyi Yang
    EMNLP (2020)
    Preview abstract We present ToTTo, an open-domain English table-to-text dataset with over 120,000 training examples that proposes a controlled generation task: given a Wikipedia table and a set of highlighted table cells, produce a one-sentence description. To obtain generated targets that are natural but also faithful to the source table, we introduce a dataset construction process where annotators directly revise existing candidate sentences from Wikipedia. We present systematic analyses of our dataset and annotation process as well as results achieved by several state-of-the-art baselines. While usually fluent, existing methods often hallucinate phrases that are not supported by the table, suggesting that this dataset can serve as a useful research benchmark for high-precision conditional text generation. View details
    Preview abstract Much of the previous machine learning (ML) fairness literature assumes that protected features such as race and sex are present in the dataset, and relies upon them to mitigate fairness concerns. However, in practice factors like privacy and regulation often preclude the collection of protected features, or their use for training or inference, severely limiting the applicability of traditional fairness research. Therefore we ask: How can we train an ML model to improve fairness when we do not even know the protected group memberships? In this work we address this problem by proposing Adversarially Reweighted Learning (ARL). In particular, we hypothesize that non-protected features and task labels are valuable for identifying fairness issues, and can be used to co-train an adversarial reweighting approach for improving fairness. Our results show that ARL improves Rawlsian Max-Min fairness, with notable AUC improvements for worst-case protected groups in multiple datasets, outperforming state-of-the-art alternatives. View details
    Underspecification Presents Challenges for Credibility in Modern Machine Learning
    Alexander Nicholas D'Amour
    Dan Moldovan
    Babak Alipanahi
    Alex Beutel
    Christina Chen
    Jon Deaton
    Jacob Eisenstein
    Shaobo Hou
    Ghassen Jerfel
    Yian Ma
    Akinori Mitani
    Andrea Montanari
    Christopher Nielsen
    Thomas Osborne
    Rajiv Raman
    Kim Ramasamy
    Rory Abbott Sayres
    Martin Gamunu Seneviratne
    Shannon Sequeira
    Harini Suresh
    Victor Veitch
    Journal of Machine Learning Research (2020)
    Preview abstract ML models often exhibit unexpectedly poor behavior when they are deployed in real-world domains. We identify underspecification as a key reason for these failures. An ML pipeline is underspecified when it can return many predictors with equivalently strong held-out performance in the training domain. Underspecification is common in modern ML pipelines, such as those based on deep learning. Predictors returned by underspecified pipelines are often treated as equivalent based on their training domain performance, but we show here that such predictors can behave very differently in deployment domains. This ambiguity can lead to instability and poor model behavior in practice, and is a distinct failure mode from previously identified issues arising from structural mismatch between training and deployment domains. We show that this problem appears in a wide variety of practical ML pipelines, using examples from computer vision, medical imaging, natural language processing, clinical risk prediction based on electronic health records, and medical genomics. Our results show the need to explicitly account for underspecification in modeling pipelines that are intended for real-world deployment in any domain. View details
    Preview abstract Large pre-trained models have revolutionized natural language understanding. However, researchers have found they can encode correlations undesired in many applications, like \emph{surgeon} being associated more with \emph{he} than \emph{she}. We explore such \emph{gendered correlations} as a case study, to learn how we can configure and train models to mitigate the risk of encoding unintended associations. We find that it is important to define correlation metrics, since they can reveal differences among models with similar accuracy. Large models have more capacity to encode gendered correlations, but this can be mitigated with general dropout regularization. Counterfactual data augmentation is also effective, and can even reduce correlations not explicitly targeted for mitigation, potentially making it useful beyond gender too. Both techniques yield models with comparable accuracy to unmitigated analogues, and still resist re-learning correlations in fine-tuning. View details
    Preview abstract NLP models are shown to suffer from robustness issues, for example, a model's prediction can be easily changed under small perturbations to the input. In this work, we aim to present a Controlled Adversarial Text Generation (CAT-Gen) model that, given an input text, it can generate adversarial texts through controllable attributes that are known to be invariant to task labels. For example, for a main task like sentiment classification, an example attribute can be different categories/domains, and a model should have similar performance across them; for a coreference resolution task, a model's performance should not differ across different demographic attributes. Different from many existing adversarial text generation approaches, we show that our model can generate adversarial texts that are more fluent, diverse, and with better task-label invariance guarantees. We aim to use this model to generate counterfactual texts that could better improve robustness in NLP models (e.g., through adversarial training), and we argue that our generation can create more natural attacks. View details
    Transfer of Machine Learning Fairness across Domains
    Candice Schumann
    Alex Beutel
    Jilin Chen
    Hai Qian
    Preview abstract If our models are used in new or unexpected cases, do we know if they will make fair predictions? Previously, researchers developed ways to debias a model for a single problem domain. However, this is often not how models are trained and used in practice. For example, labels and demographics (sensitive attributes) are often hard to observe, resulting in auxiliary or synthetic data to be used for training, and proxies of the sensitive attribute to be used for evaluation of fairness. A model trained for one setting may be picked up and used in many others, particularly as is common with pre-training and cloud APIs. Despite the pervasiveness of these complexities, remarkably little work in the fairness literature has theoretically examined these issues. We frame all of these settings as domain adaptation problems: how can we use what we have learned in a source domain to debias in a new target domain, without directly debiasing on the target domain as if it is a completely new problem? We offer new theoretical guarantees of improving fairness across domains, and offer a modeling approach to transfer to data-sparse target domains. We give empirical results validating the theory and showing that these modeling approaches can improve fairness metrics with less data. View details
    Preview abstract The launch of the new Google News in 2018 introduced the Frequently asked questions feature to structurally summarize the news story in its full coverage page. While news summarization has been a research topic for decades, this new feature is poised to usher in a new line of news summarization techniques. There are two fundamental approaches: mining the questions from data associated with the news story and learning the questions from the content of the story directly. This paper provides the first study, to the best of our knowledge, of a learning based approach to generate a structured summary of news articles with question and answer pairs to capture salient and interesting aspects of the news story. Specifically, this learning-based approach reads a news article, predicts its attention map (i.e., important snippets in the article), and generates multiple natural language questions corresponding to each snippet. Furthermore, we describe a mining-based approach as the mechanism to generate weak supervision data for training the learning based approach. We evaluate our approach on the existing SQuAD dataset2 and a large dataset with 91K news articles we constructed. We show that our proposed system can achieve an AUC of 0:734 for document attention map prediction, a BLEU-4 score of 12:46 for natural question generation and a BLEU-4 score of 24:4 for question summarization, beating state-of-art baselines. View details
    Verifying Text Summaries of Relational Data Sets
    Saehan Jo
    Immanuel Trummer
    Weicheng Yu
    Cong Yu
    Daniel Liu
    Niyati Mehta
    SIGMOD (2019) (to appear)
    Preview abstract We present a novel natural language query interface, the AggChecker, aimed at text summaries of relational data sets. The tool focuses on natural language claims that translate into an SQL query and a claimed query result. Similar in spirit to a spell checker, the AggChecker marks up text passages that seem to be inconsistent with the actual data. At the heart of the system is a probabilistic model that reasons about the input document in a holistic fashion. Based on claim keywords and the document structure, it maps each text claim to a probability distribution over associated query translations. By efficiently executing tens to hundreds of thousands of candidate translations for a typical input document, the system maps text claims to correctness probabilities. This process becomes practical via a specialized processing backend, avoiding redundant work via query merging and result caching. Verification is an interactive process in which users are shown tentative results, enabling them to take corrective actions if necessary. Our system was tested on a set of 53 public articles containing 392 claims. Our test cases include articles from major newspapers, summaries of survey results, and Wikipedia articles. Our tool revealed erroneous claims in roughly a third of test cases. A detailed user study shows that users using our tool are in average six times faster at checking text summaries, compared to generic SQL interfaces. In fully automated verification, our tool achieves significantly higher recall and precision than baselines from the areas of natural language query interfaces and fact-checking. View details
    Preview abstract Modern search engines increasingly incorporate tabular content, which consists of a set of entities each augmented with a small set of facts. The facts can be obtained from multiple sources: an entity’s knowledge base entry, the infobox on its Wikipedia page, or its row within a WebTable. Crucially, the informativeness of a fact depends not only on the entity but also the specific context (e.g., the query). To the best of our knowledge, this paper is the first to study the problem of contextual fact ranking: given some entities and a con- text (i.e., succinct natural language description), identify the most informative facts for the entities collectively within the context. We propose to contextually rank the facts by exploiting deep learning techniques. In particular, we develop pointwise and pair- wise ranking models, using textual and statistical information for the given entities and context derived from their sources. We en- hance the models by incorporating entity type information from an IsA (hypernym) database. We demonstrate that our approaches achieve better performance than state-of-the-art baselines in terms of MAP, NDCG, and recall. We further conduct user studies for two specific applications of contextual fact ranking—table synthesis and table compression—and show that our models can identify more informative facts than the baselines. View details
    Preview abstract Modern search engines provide contextual information surrounding query entities beyond ``ten blue links'' in the form of knowledge cards. Among the various attributes displayed about entities there has been recent interest in providing trivia due to observed engagement rates. Obtaining such trivia at a large scale is, however, non-trivial: hiring professional content creators is expensive and extracting statements from the Web can result in unreliable or uninteresting facts. In this paper we show how fun facts can be mined from tables on the Web to provide a large volume of reliable and interesting content. We employ a template-based approach to generate statements that are postprocessed by workers. We show how to bootstrap and streamline the process for faster and cheaper task completion. However, the content contained in these tables is dynamic. Therefore, we address the problem of automatically maintaining templates when tables are updated. View details
    AggChecker: A Fact-Checking System for Text Summaries of Relational Data Sets
    Saehan Jo
    Immanuel Trummer
    Weicheng Yu
    Cong Yu
    Daniel Liu
    Niyati Mehta
    VLDB (2019)
    Preview abstract We demonstrate AggChecker, a novel tool for verifying textual summaries of relational data sets. The system automatically verifies natural language claims about numerical aggregates against the underlying raw data. The system incorporates a combination of natural language processing, information retrieval, machine learning, and efficient query processing strategies. Each claim is translated into a semantically equivalent SQL query and evaluated against the database. Our primary goal is analogous to that of a spell-checker: to identify erroneous claims and provide guidance in correcting them. In this demonstration, we show that our system enables users to verify text summaries much more efficiently than a standard SQL interface. View details
    Mining an “Anti-Knowledge Base” from Wikipedia Updates with Applications to Fact Checking and Beyond
    Georgios Karagiannis
    Immanuel Trummer
    Saehan Jo
    Shubham Khandelwal
    Cong Yu
    VLDB 2019
    Preview abstract We introduce the problem of anti-knowledge mining. Our goal is to create an “anti-knowledge base” that contains factual mistakes. The resulting data can be used for analysis, training, and benchmarking in the research domain of auto-mated fact checking. Prior data sets feature manually generated fact checks of famous misclaims. Instead, we focus on the long tail of factual mistakes made by Web authors, ranging from erroneous sports results to incorrect capitals. We mine mistakes automatically, by an unsupervised approach, from Wikipedia updates that correct factual mistakes. Identifying such updates (only a small fraction of the total number of updates) is one of the primary challenges. We mine anti-knowledge by a multi-step pipeline. First, we filter out candidate updates via several simple heuristics. Next, we correlate Wikipedia updates with other statements made on the Web. Using claim occurrence frequencies as input to a probabilistic model, we infer the likelihood of corrections via an iterative expectation-maximization approach. Finally, we extract mistakes in the form of subject-predicate-object triples and rank them according to several criteria. Our end result is a data set containing over 110,000 ranked mistakes with a precision of 85% in the top 1% and a precision of over 60% in the top 25%. We demonstrate that baselines achieve significantly lower precision. Also, we exploit our data to verify several hypothesis on why users make mistakes. We finally show that the AKB can be used to find mistakes on the entire Web. View details
    Preview abstract With the support of major search platforms such as Google and Bing, fact-checking articles, which can be identified by their adoption of the schema.org ClaimReview structured markup, have gained widespread recognition for their role in the fight against digital misinformation. A claim-relevant document is an online document that addresses, and potentially expresses a stance towards, some claim. The claim-relevance discovery problem, then, is to find claim-relevant documents. Depending on the verdict from the fact check, claim-relevance discovery can help identify online misinformation. In this paper, we provide an initial approach to the claim-relevance discovery problem by leveraging various information retrieval and machine learning techniques. The system consists of three phases. First, we retrieve candidate documents based on various features in the fact-checking article. Second, we apply a relevance classifier to filter away documents that do not address the claim. Third, we apply a language feature based classifier to distinguish documents with different stances towards the claim. We experimentally demonstrate that our solution achieves solid results on a large-scale dataset and beats state-of-the-art baselines. Finally, we highlight a rich set of case studies to demonstrate the myriad of remaining challenges and that this problem is far from being solved. View details
    Biperpedia: An Ontology for Search Applications
    Alon Halevy
    Steven Whang
    Fei Wu
    Proc. 40th Int'l Conf. on Very Large Data Bases (PVLDB) (2014)
    Preview abstract Search engines make significant efforts to recognize queries that can be answered by structured data and invest heavily in creating and maintaining high-precision databases. While these databases have a relatively wide coverage of entities, the number of attributes they model (e.g., gdp, capital, anthem) is relatively small. Extending the number of attributes known to the search engine can enable it to more precisely answer queries from the long and heavy tail, extract a broader range of facts from the Web, and recover the semantics of tables on the Web. We describe Biperpedia, an ontology with 1.6M (class, attribute) pairs and 67K distinct attribute names. Biperpedia extracts attributes from the query stream, and then uses the best extractions to seed attribute extraction from text. For every attribute Biperpedia saves a set of synonyms and text patterns in which it appears, thereby enabling it to recognize the attribute in more contexts. In addition to a detailed analysis of the quality of Biperpedia, we show that it can increase the number of Web tables whose semantics we can recover by more than a factor of 4 compared with Freebase. View details
    No Results Found