Machine Reading at Scale (MRS) is a challenging task in which a system is given an input query and is asked to produce a precise output by "reading" information from a large knowledge base. The task has gained popularity with its natural combination of information retrieval (IR) and machine comprehension (MC). Advancements in representation learning have led to separated progress in both IR and MC; however, very few studies have examined the relationship and combined design of retrieval and comprehension at different levels of granularity, for development of MRS systems. In this work, we give general guidelines on system design for MRS by proposing a simple yet effective pipeline system with special consideration on hierarchical semantic retrieval at both paragraph and sentence level, and their potential effects on the downstream task. The system is evaluated on both fact verification and open-domain multihop QA, achieving state-of-the-art results on the leaderboard test sets of both FEVER and HOTPOTQA. To further demonstrate the importance of semantic retrieval, we present ablation and analysis studies to quantify the contribution of neural retrieval modules at both paragraph-level and sentence-level, and illustrate that intermediate semantic retrieval modules are vital for not only effectively filtering upstream information and thus saving downstream computation, but also for shaping upstream data distribution and providing better data for downstream modeling. Code/data made publicly available at: https://github.com/easonnie/semanticRetrievalMRS
This paper describes the participation of UvA.ILPS group at the TREC CAsT 2020 track. Our passage retrieval pipeline consists of (i) an initial retrieval module that uses BM25, and (ii) a re-ranking module that combines the score of a BERT ranking model with the score of a machine comprehension model adjusted for passage retrieval. An important challenge in conversational passage retrieval is that queries are often under-specified. Thus, we perform query resolution, that is, add missing context from the conversation history to the current turn query using QuReTeC, a term classification query r…
Carnegie Mellon University（卡内基梅隆大学）；Bosch Research North America（博世北美研究院）；Amazon（亚马逊）
Systems for knowledge-intensive tasks such as open-domain question answering (QA) usually consist of two stages: efficient retrieval of relevant documents from a large corpus and detailed reading of the selected documents to generate answers. Retrievers and readers are usually modeled separately, which necessitates a cumbersome implementation and is hard to train and adapt in an end-to-end fashion. In this paper, we revisit this design and eschew the separate architecture and training in favor of a single Transformer that performs Retrieval as Attention (ReAtt), and end-to-end training solely …
Denison University（丹尼森大学）；Vietnam National University, Ho Chi Minh City（胡志明市越南国家大学）；University of Information Technology, Ho Chi Minh City（胡志明市信息技术大学）
Although the curse of multilinguality significantly restricts the language abilities of multilingual models in monolingual settings, researchers now still have to rely on multilingual models to develop state-of-the-art systems in Vietnamese Machine Reading Comprehension. This difficulty in researching is because of the limited number of high-quality works in developing Vietnamese language models. In order to encourage more work in this research field, we present a comprehensive analysis of language weaknesses and strengths of current Vietnamese monolingual models using the downstream task of M…
Tsinghua University（清华大学）；Renmin University of China（中国人民大学）；BNRist（北京国家信息科学与技术研究中心）；Institute for Artificial Intelligence, Tsinghua University（清华大学人工智能研究院）
Knowledge graphs, as the cornerstone of many AI applications, usually face serious incompleteness problems. In recent years, there have been many efforts to study automatic knowledge graph completion (KGC), most of which use existing knowledge to infer new knowledge. However, in our experiments, we find that not all relations can be obtained by inference, which constrains the performance of existing models. To alleviate this problem, we propose a new model based on information retrieval and reading comprehension, namely IR4KGC. Specifically, we pre-train a knowledge-based information retrieval…
NTT Media Intelligence Laboratories, NTT Corporation（日本电信电话株式会社（NTT）媒体智能实验室）
This study considers the task of machine reading at scale (MRS) wherein, given a question, a system first performs the information retrieval (IR) task of finding relevant passages in a knowledge source and then carries out the reading comprehension (RC) task of extracting an answer span from the passages. Previous MRS studies, in which the IR component was trained without considering answer spans, struggled to accurately find a small number of relevant passages from a large set of passages. In this paper, we propose a simple and effective approach that incorporates the IR and RC tasks by using…
Pre-trained Language Models (PLMs) have achieved great success on Machine Reading Comprehension (MRC) over the past few years. Although the general language representation learned from large-scale corpora does benefit MRC, the poor support in evidence extraction which requires reasoning across multiple sentences hinders PLMs from further advancing MRC. To bridge the gap between general PLMs and MRC, we present REPT, a REtrieval-based Pre-Training approach. In particular, we introduce two self-supervised tasks to strengthen evidence extraction during pre-training, which is further inherited by …
Beijing Institute of Technology（北京理工大学）；Southeast Academy of Information Technology（东南信息技术研究院）
Open-retrieval conversational machine reading comprehension (OCMRC) simulates real-life conversational interaction scenes. Machines are required to make a decision of "Yes/No/Inquire" or generate a follow-up question when the decision is "Inquire" based on retrieved rule texts, user scenario, user question, and dialogue history. Recent studies explored the methods to reduce the information gap between decision-making and question generation and thus improve the performance of generation. However, the information gap still exists because these pipeline structures are still limited in decision-m…
University of Washington（华盛顿大学）；Princeton University（普林斯顿大学）；Facebook AI Research（Facebook人工智能研究院）；Allen Institute for AI（艾伦人工智能研究所）
We introduce an approach for open-domain question answering (QA) that retrieves and reads a passage graph, where vertices are passages of text and edges represent relationships that are derived from an external knowledge base or co-occurrence in the same article. Our goals are to boost coverage by using knowledge-guided retrieval to find more relevant passages than text-matching methods, and to improve accuracy by allowing for better knowledge-guided fusion of information across related passages. Our graph retrieval method expands a set of seed keyword-retrieved passages by traversing the grap…
University of Amsterdam（阿姆斯特丹大学）；Amazon Alexa AI（亚马逊Alexa人工智能部门）
Interacting with a speech interface to query a Question Answering (QA) system is becoming increasingly popular. Typically, QA systems rely on passage retrieval to select candidate contexts and reading comprehension to extract the final answer. While there has been some attention to improving the reading comprehension part of QA systems against errors that automatic speech recognition (ASR) models introduce, the passage retrieval part remains unexplored. However, such errors can affect the performance of passage retrieval, leading to inferior end-to-end performance. To address this gap, we augm…
In recent years, there have been amazing advances in deep learning methods for machine reading. In machine reading, the machine reader has to extract the answer from the given ground truth paragraph. Recently, the state-of-the-art machine reading models achieve human level performance in SQuAD which is a reading comprehension-style question answering (QA) task. The success of machine reading has inspired researchers to combine information retrieval with machine reading to tackle open-domain QA. However, these systems perform poorly compared to reading comprehension-style QA because it is diffi…
ETH Zürich（苏黎世联邦理工学院）；University of Massachusetts Amherst（马萨诸塞大学阿默斯特分校）
In typical machine learning systems, an estimate of the probability of the prediction is used to assess the system's confidence in the prediction. This confidence measure is usually uncalibrated; i.e.\ the system's confidence in the prediction does not match the true probability of the predicted output. In this paper, we present an investigation into calibrating open setting machine reading systems such as open-domain question answering and claim verification systems. We show that calibrating such complex systems which contain discrete retrieval and deep reading components is challenging and c…
Indian Institute of Technology, Kanpur（印度理工学院坎普尔分校）；Vellore Institute of Technology（韦洛尔理工学院）
This paper describes our submission to the Competition on Legal Information Extraction/Entailment 2022 (COLIEE-2022) workshop on case law competition for tasks 1 and 2. Task 1 is a legal case retrieval task, which involves reading a new case and extracting supporting cases from the provided case law corpus to support the decision. Task 2 is the legal case entailment task, which involves the identification of a paragraph from existing cases that entails the decision in a relevant case. We employed the neural models Sentence-BERT and Sent2Vec for semantic understanding and the traditional retrie…
School of Computer Science（计算机科学学院）；Carnegie Mellon University（卡内基梅隆大学）
We present two new large-scale datasets aimed at evaluating systems designed to comprehend a natural language query and extract its answer from a large corpus of text. The Quasar-S dataset consists of 37000 cloze-style (fill-in-the-gap) queries constructed from definitions of software entity tags on the popular website Stack Overflow. The posts and comments on the website serve as the background corpus for answering the cloze questions. The Quasar-T dataset consists of 43000 open-domain trivia questions and their answers obtained from various internet sources. ClueWeb09 serves as the backgroun…
Information Extraction (IE) from text refers to the task of extracting structured knowledge from unstructured text. The task typically consists of a series of sub-tasks such as Named Entity Recognition and Relation Extraction. Sourcing entity and relation type specific training data is a major bottleneck in domains with limited resources such as biomedicine. In this work we present a slot filling approach to the task of biomedical IE, effectively replacing the need for entity and relation-specific training data, allowing us to deal with zero-shot settings. We follow the recently proposed parad…
Academia Sinica（中央研究院）；National Taiwan University（国立台湾大学）
In recent years, the number of people studying English as a second language (ESL) has surpassed the number of native speakers. Recent work have demonstrated the success of providing personalized content based on reading difficulty, such as information retrieval and summarization. However, almost all prior studies of reading difficulty are designed for native speakers, rather than non-native readers. In this study, we investigate various features for ESL readers, by conducting a linear regression to estimate the reading level of English language sources. This estimation is based not only on the…
Stanford University（斯坦福大学）；Facebook AI Research（Facebook人工智能研究院）
This paper proposes to tackle open- domain question answering using Wikipedia as the unique knowledge source: the answer to any factoid question is a text span in a Wikipedia article. This task of machine reading at scale combines the challenges of document retrieval (finding the relevant articles) with that of machine comprehension of text (identifying the answer spans from those articles). Our approach combines a search component based on bigram hashing and TF-IDF matching with a multi-layer recurrent neural network model trained to detect answers in Wikipedia paragraphs. Our experiments on …
University of Arizona（亚利桑那大学）；IBM Research（IBM研究院）
We propose a simple and effective strategy for data augmentation for low-resource machine reading comprehension (MRC). Our approach first pretrains the answer extraction components of a MRC system on the augmented data that contains approximate context of the correct answers, before training it on the exact answer spans. The approximate context helps the QA method components in narrowing the location of the answers. We demonstrate that our simple strategy substantially improves both document retrieval and answer extraction performance by providing larger context of the answers and additional t…
Reading comprehension models are based on recurrent neural networks that sequentially process the document tokens. As interest turns to answering more complex questions over longer documents, sequential reading of large portions of text becomes a substantial bottleneck. Inspired by how humans use document structure, we propose a novel framework for reading comprehension. We represent documents as trees, and model an agent that learns to interleave quick navigation through the document tree with more expensive answer extraction. To encourage exploration of the document tree, we propose a new al…
A major challenge of research on non-English machine reading for question answering (QA) is the lack of annotated datasets. In this paper, we present GermanQuAD, a dataset of 13,722 extractive question/answer pairs. To improve the reproducibility of the dataset creation approach and foster QA research on other languages, we summarize lessons learned and evaluate reformulation of question/answer pairs as a way to speed up the annotation process. An extractive QA model trained on GermanQuAD significantly outperforms multilingual models and also shows that machine-translated training data cannot …
University of Illinois at Urbana-Champaign（伊利诺伊大学厄巴纳-香槟分校）；IBM Research AI, New York（IBM研究院人工智能实验室，纽约）
Recent work has shown that commonly available machine reading comprehension (MRC) datasets can be used to train high-performance neural information retrieval (IR) systems. However, the evaluation of neural IR has so far been limited to standard supervised learning settings, where they have outperformed traditional term matching baselines. We conduct in-domain and out-of-domain evaluations of neural IR, and seek to improve its robustness across different scenarios, including zero-shot settings. We show that synthetic training examples generated using a sequence-to-sequence generator can be effe…
The University of Arizona（亚利桑那大学）；Allen Institute for Artificial Intelligence（艾伦人工智能研究所）
Considerable progress has been made recently in open-domain question answering (QA) problems, which require Information Retrieval (IR) and Reading Comprehension (RC). A popular approach to improve the system's performance is to improve the quality of the retrieved context from the IR stage. In this work we show that for StrategyQA, a challenging open-domain QA dataset that requires multi-hop reasoning, this common approach is surprisingly ineffective -- improving the quality of the retrieved context hardly improves the system's performance. We further analyze the system's behavior to identify …
University of Arkansas（阿肯色大学）；Google LLC.（谷歌有限责任公司）
Deep language models learning a hierarchical representation proved to be a powerful tool for natural language processing, text mining and information retrieval. However, representations that perform well for retrieval must capture semantic meaning at different levels of abstraction or context-scopes. In this paper, we propose a new method to generate multi-resolution word embeddings that represent documents at multiple resolutions in terms of context-scopes. In order to investigate its performance,we use the Stanford Question Answering Dataset (SQuAD) and the Question Answering by Search And R…
University of Amsterdam（阿姆斯特丹大学）；Utrecht University（乌得勒支大学）
In many real-world scenarios, the absence of external knowledge source like Wikipedia restricts question answering systems to rely on latent internal knowledge in limited dialogue data. In addition, humans often seek answers by asking several questions for more comprehensive information. As the dialog becomes more extensive, machines are challenged to refer to previous conversation rounds to answer questions. In this work, we propose to leverage latent knowledge in existing conversation logs via a neural Retrieval-Reading system, enhanced with a TFIDF-based text summarizer refining lengthy con…
Reading comprehension (RC)---in contrast to information retrieval---requires integrating information and reasoning about events, entities, and their relations across a full document. Question answering is conventionally used to assess RC ability, in both artificial agents and children learning to read. However, existing RC datasets and tasks are dominated by questions that can be solved by selecting answers using superficial information (e.g., local context similarity or global term frequency); they thus fail to test for the essential integrative aspect of RC. To encourage progress on deeper c…
Wangxuan Institute of Computer Technology, Peking University（北京大学王选计算机研究所）；Peking University（北京大学）
Recent studies report that many machine reading comprehension (MRC) models can perform closely to or even better than humans on benchmark datasets. However, existing works indicate that many MRC models may learn shortcuts to outwit these benchmarks, but the performance is unsatisfactory in real-world applications. In this work, we attempt to explore, instead of the expected comprehension skills, why these models learn the shortcuts. Based on the observation that a large portion of questions in current datasets have shortcut solutions, we argue that larger proportion of shortcut questions in tr…
This paper tackles the problem of open domain factual Arabic question answering (QA) using Wikipedia as our knowledge source. This constrains the answer of any question to be a span of text in Wikipedia. Open domain QA for Arabic entails three challenges: annotated QA datasets in Arabic, large scale efficient information retrieval and machine reading comprehension. To deal with the lack of Arabic QA datasets we present the Arabic Reading Comprehension Dataset (ARCD) composed of 1,395 questions posed by crowdworkers on Wikipedia articles, and a machine translation of the Stanford Question Answe…
A fundamental trade-off between effectiveness and efficiency needs to be balanced when designing an online question answering system. Effectiveness comes from sophisticated functions such as extractive machine reading comprehension (MRC), while efficiency is obtained from improvements in preliminary retrieval components such as candidate document selection and paragraph ranking. Given the complexity of the real-world multi-document MRC scenario, it is difficult to jointly optimize both in an end-to-end system. To address this problem, we develop a novel deep cascade learning model, which progr…
Sofia University（索非亚大学）；Hamad Bin Khalifa University（哈马德·本·哈利法大学）；Qatar Computing Research Institute（卡塔尔计算研究所）
Recent advances in deep neural networks, language modeling and language generation have introduced new ideas to the field of conversational agents. As a result, deep neural models such as sequence-to-sequence, Memory Networks, and the Transformer have become key ingredients of state-of-the-art dialog systems. While those models are able to generate meaningful responses even in unseen situation, they need a lot of training data to build a reliable model. Thus, most real-world systems stuck to traditional approaches based on information retrieval and even hand-crafted rules, due to their robustn…
Recent advances in conversational systems have changed the search paradigm. Traditionally, a user poses a query to a search engine that returns an answer based on its index, possibly leveraging external knowledge bases and conditioning the response on earlier interactions in the search session. In a natural conversation, there is an additional source of information to take into account: utterances produced earlier in a conversation can also be referred to and a conversational IR system has to keep track of information conveyed by the user during the conversation, even if it is implicit. We arg…
The Epidemic Question Answering (EPIC-QA) track at the Text Analysis Conference (TAC) is an evaluation of methodologies for answering ad-hoc questions about the COVID-19 disease. This paper describes our participation in both tasks of EPIC-QA, targeting: (1) Expert QA and (2) Consumer QA. Our methods used a multi-phase neural Information Retrieval (IR) system based on combining BM25, BERT, and T5 as well as the idea of considering entailment relations between the original question and questions automatically generated from answer candidate sentences. Moreover, because entailment relations were…
Chinese University of Hong Kong（香港中文大学）；Salesforce Research（Salesforce研究院）
In conversational machine reading, systems need to interpret natural language rules, answer high-level questions such as "May I qualify for VA health care benefits?", and ask follow-up clarification questions whose answer is necessary to answer the original question. However, existing works assume the rule text is provided for each user question, which neglects the essential retrieval step in real scenarios. In this work, we propose and investigate an open-retrieval setting of conversational machine reading. In the open-retrieval setting, the relevant rule texts are unknown so that a system ne…
University of North Carolina at Chapel Hill（北卡罗来纳大学教堂山分校）
Training an effective video-and-language model intuitively requires multiple frames as model inputs. However, it is unclear whether using multiple frames is beneficial to downstream tasks, and if yes, whether the performance gain is worth the drastically-increased computation and memory costs resulting from using more frames. In this work, we explore single-frame models for video-and-language learning. On a diverse set of video-and-language tasks (including text-to-video retrieval and video question answering), we show the surprising result that, with large-scale pre-training and a proper fram…
Facebook AI Research（脸书人工智能研究院）；University College London（伦敦大学学院）
When pre-trained on large unsupervised textual corpora, language models are able to store and retrieve factual knowledge to some extent, making it possible to use them directly for zero-shot cloze-style question answering. However, storing factual knowledge in a fixed number of weights of a language model clearly has limitations. Previous approaches have successfully provided access to information outside the model weights using supervised architectures that combine an information retrieval system with a machine reading component. In this paper, we go a step further and integrate information f…
MoM: Mixtures of Scenario-Aware Document Memories for Retrieval-Augmented Generation Systems
School of Information, Renmin University of China, Beijing, China（中国人民大学信息学院）；MemTensor (Shanghai) Technology Co., Ltd.（MemTensor（上海）技术有限公司）；Institute for Advanced Algorithms Research, Shanghai（上海先进算法研究所）
DeepRead: Document Structure-Aware Reasoning to Enhance Agentic Search
State Key Laboratory of AI Safety, Institute of Computing Technology, Chinese Academy of Sciences (CAS)（人工智能安全国家重点实验室，计算技术研究所，中国科学院）；University of Chinese Academy of Sciences, CAS（中国科学院大学）；Wenlan School of Business, Zhongnan University of Economics（中南财经政法大学文澜商学院）
State Key Laboratory for Novel Software Technology, Nanjing University（南京大学计算机软件新技术国家重点实验室）
Scenario-based question answering (SQA) requires retrieving and reading paragraphs from a large corpus to answer a question which is contextualized by a long scenario description. Since a scenario contains both keyphrases for retrieval and much noise, retrieval for SQA is extremely difficult. Moreover, it can hardly be supervised due to the lack of relevance labels of paragraphs for SQA. To meet the challenge, in this paper we propose a joint retriever-reader model called JEEVES where the retriever is implicitly supervised only using QA labels via a novel word weighting mechanism. JEEVES signi…
Bergische Universität Wuppertal（伍珀塔尔大学）；Universität Hamburg（汉堡大学）
The corpus, from which a predictive language model is trained, can be considered the experience of a semantic system. We recorded everyday reading of two participants for two months on a tablet, generating individual corpus samples of 300/500K tokens. Then we trained word2vec models from individual corpora and a 70 million-sentence newspaper corpus to obtain individual and norm-based long-term memory structure. To test whether individual corpora can make better predictions for a cognitive task of long-term memory retrieval, we generated stimulus materials consisting of 134 sentences with uncor…
This paper presents the "Leipzig Corpus Miner", a technical infrastructure for supporting qualitative and quantitative content analysis. The infrastructure aims at the integration of 'close reading' procedures on individual documents with procedures of 'distant reading', e.g. lexical characteristics of large document collections. Therefore information retrieval systems, lexicometric statistics and machine learning procedures are combined in a coherent framework which enables qualitative data analysts to make use of state-of-the-art Natural Language Processing techniques on very large document …
Benchmarks for Pirá 2.0, a Reading Comprehension Dataset about the Ocean, the Brazilian Coast, and Climate Change
Universidade de São Paulo（圣保罗大学）；Instituto de Estudos Avançados（高等研究院）；Escola Politécnica（理工学院）；Instituto de Matemática e Estatística（数学与统计研究所）；Escola de Artes, Ciências e Humanidades（艺术、科学与人文学院）
Google Research（谷歌研究院）；Paul G. Allen School of Computer Science & Engineering, University of Washington（华盛顿大学保罗·G·艾伦计算机科学与工程学院）
Deep NLP models have been shown to learn spurious correlations, leaving them brittle to input perturbations. Recent work has shown that counterfactual or contrastive data -- i.e. minimally perturbed inputs -- can reveal these weaknesses, and that data augmentation using counterfactuals can help ameliorate them. Proposed techniques for generating counterfactuals rely on human annotations, perturbations based on simple heuristics, and meaning representation frameworks. We focus on the task of creating counterfactuals for question answering, which presents unique challenges related to world knowl…
SenseTime（商汤科技）；National University of Defense Technology（国防科技大学）；Singapore Management University（新加坡管理大学）
Document-level natural language inference (DOCNLI) is a new challenging task in natural language processing, aiming at judging the entailment relationship between a pair of hypothesis and premise documents. Current datasets and baselines largely follow sentence-level settings, but fail to address the issues raised by longer documents. In this paper, we establish a general solution, named Retrieval, Reading and Fusion (R2F) framework, and a new setting, by analyzing the main challenges of DOCNLI: interpretability, long-range dependency, and cross-sentence inference. The basic idea of the framew…
Transformer-based large language models are trained to make predictions about the next word by aggregating representations of previous tokens through their self-attention mechanism. In the field of cognitive modeling, such attention patterns have recently been interpreted as embodying the process of cue-based retrieval, in which attention over multiple targets is taken to generate interference and latency during retrieval. Under this framework, this work first defines an entropy-based predictor that quantifies the diffuseness of self-attention, as well as distance-based predictors that capture…
University of North Carolina Chapel Hill（北卡罗来纳大学教堂山分校）；Johns Hopkins University（约翰斯·霍普金斯大学）
Answering complex questions often requires multi-step reasoning in order to obtain the final answer. Most research into decompositions of complex questions involves open-domain systems, which have shown success in using these decompositions for improved retrieval. In the machine reading setting, however, work to understand when decompositions are helpful is understudied. We conduct experiments on decompositions in machine reading to unify recent work in this space, using a range of models and datasets. We find that decompositions can be helpful in the few-shot case, giving several points of im…
