Filters








24 Hits in 3.5 sec

XOR QA: Cross-lingual Open-Retrieval Question Answering [article]

Akari Asai, Jungo Kasai, Jonathan H. Clark, Kenton Lee, Eunsol Choi, Hannaneh Hajishirzi
2021 arXiv   pre-print
Our task formulation, called Cross-lingual Open Retrieval Question Answering (XOR QA), includes 40k information-seeking questions from across 7 diverse non-English languages.  ...  This work extends open-retrieval question answering to a cross-lingual setting enabling questions from one language to be answered via answer content from another language.  ...  XOR-FULL defines our goal of building a multilingual open-retrieval QA system that uses both cross-lingual and in-language questions from XOR-TYDI QA.  ... 
arXiv:2010.11856v3 fatcat:7mujquejk5fxxmhggc22plic7u

Addressing Issues of Cross-Linguality in Open-Retrieval Question Answering Systems For Emergent Domains [article]

Alon Albalak, Sharon Levy, William Yang Wang
2022 arXiv   pre-print
In this paper, we demonstrate a cross-lingual open-retrieval question answering system for the emergent domain of COVID-19.  ...  Open-retrieval question answering systems are generally trained and tested on large datasets in well-established domains.  ...  This system demonstration provides in-depth technical descriptions of the individual components of our cross-lingual open-retrieval question answering system: cross-lingual retrieval and crosslingual reading  ... 
arXiv:2201.11153v1 fatcat:ion7747l4vb6nmkwnk3hsul76e

XOR QA: Cross-lingual Open-Retrieval Question Answering

Akari Asai, Jungo Kasai, Jonathan Clark, Kenton Lee, Eunsol Choi, Hannaneh Hajishirzi
2021 Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies   unpublished
Based on this dataset, we introduce a task framework, called Cross-lingual Open-Retrieval Question Answering (XOR QA), that consists of three new tasks involving crosslingual document retrieval from multilingual  ...  This work extends open-retrieval question answering to a cross-lingual setting enabling questions from one language to be answered via answer content from another language.  ...  XOR-FULL defines our goal of building a multilingual open-retrieval QA system that uses both cross-lingual and in-language questions from XOR-TYDI QA.  ... 
doi:10.18653/v1/2021.naacl-main.46 fatcat:czlwansf65bltdcp7bie4haske

SD-QA: Spoken Dialectal Question Answering for the Real World [article]

Fahim Faisal, Sharlina Keshava, Md Mahfuz ibn Alam, Antonios Anastasopoulos
2021 arXiv   pre-print
Question answering (QA) systems are now available through numerous commercial applications for a wide variety of domains, serving millions of users that interact with them via speech interfaces.  ...  To address this gap, we augment an existing QA dataset to construct a multi-dialect, spoken QA benchmark on five languages (Arabic, Bengali, English, Kiswahili, Korean) with more than 68k audio prompts  ...  XOR QA: Cross-lingual open-retrieval question answering. arXiv:2010.11856. Douglas Bates, Martin Mächler, Ben Bolker, and Steve Walker. 2015. Fitting linear mixed-effects models using lme4.  ... 
arXiv:2109.12072v1 fatcat:pyy66fqphjb3fnw2lsk52eeovq

QA Dataset Explosion: A Taxonomy of NLP Resources for Question Answering and Reading Comprehension [article]

Anna Rogers, Matt Gardner, Isabelle Augenstein
2021 arXiv   pre-print
We further discuss the current classifications of "reasoning types" in question answering and propose a new taxonomy.  ...  Question answering and reading comprehension have been particularly prolific in this regard, with over 80 new datasets appearing in the past two years.  ...  , Japanese, Indonesian, Kiswahili, Korean, Russian, Telugu, and Thai. • XOR QA [11] builds on Tidy QA data to pose the task of cross-lingual QA: answering questions, where the answer data is unavailable  ... 
arXiv:2107.12708v1 fatcat:sfwmrimlgfg4xkmmca6wspec7i

One Question Answering Model for Many Languages with Cross-lingual Dense Passage Retrieval [article]

Akari Asai, Xinyan Yu, Jungo Kasai, Hannaneh Hajishirzi
2021 arXiv   pre-print
We present Cross-lingual Open-Retrieval Answer Generation (CORA), the first unified many-to-many question answering (QA) model that can answer questions across many languages, even for ones without language-specific  ...  Our analyses show the significance of cross-lingual retrieval and generation in many languages, particularly under low-resource settings.  ...  This work focuses on multilingual open QA, which involves not only machine reading comprehension but also cross-lingual retrieval.  ... 
arXiv:2107.11976v2 fatcat:fhlc373mpfc3bcthq7g72vssem

Mr. TyDi: A Multi-lingual Benchmark for Dense Retrieval [article]

Xinyu Zhang, Xueguang Ma, Peng Shi, Jimmy Lin
2021 arXiv   pre-print
TyDi, a multi-lingual benchmark dataset for mono-lingual retrieval in eleven typologically diverse languages, designed to evaluate ranking with learned dense representations.  ...  In addition to analyses of our results, we also discuss future challenges and present a research agenda in multi-lingual dense retrieval. Mr.  ...  Many multi-lingual (both mono-lingual and cross-lingual) information retrieval and question answering datasets have been constructed over the past decades, via community-wide evaluations at TREC, 1 FIRE  ... 
arXiv:2108.08787v2 fatcat:4ivjwnehcbayxpmoin3y4evgvy

MKQA: A Linguistically Diverse Benchmark for Multilingual Open Domain Question Answering [article]

Shayne Longpre, Yi Lu, Joachim Daiber
2021 arXiv   pre-print
Progress in cross-lingual modeling depends on challenging, realistic, and diverse evaluation sets.  ...  We introduce Multilingual Knowledge Questions and Answers (MKQA), an open-domain question answering evaluation set comprising 10k question-answer pairs aligned across 26 typologically diverse languages  ...  Xor-QA (Asai et al., 2021) explores cross-lingual subtasks by re-annotating 40k TyDi examples, over 7 languages, sourcing answers from English documents and translating them back to the target language  ... 
arXiv:2007.15207v2 fatcat:pmnbniyko5cjzbgv5o4hvjhpta

MKQA: A Linguistically Diverse Benchmark for Multilingual Open Domain Question Answering

Shayne Longpre, Yi Lu, Joachim Daiber
2021 Transactions of the Association for Computational Linguistics  
Progress in cross-lingual modeling depends on challenging, realistic, and diverse evaluation sets.  ...  We introduce Multilingual Knowledge Questions and Answers (MKQA), an open- domain question answering evaluation set comprising 10k question-answer pairs aligned across 26 typologically diverse languages  ...  Xor-QA (Asai et al., 2021) explores cross-lingual subtasks by re-annotating 40k TyDi examples, over 7 languages, sourcing answers from English documents and translating them back to the target language  ... 
doi:10.1162/tacl_a_00433 fatcat:hef2fubo7jfjjftxt6juo5uuf4

Pivot Through English: Reliably Answering Multilingual Questions without Document Retrieval [article]

Ivan Montero, Shayne Longpre, Ni Lao, Andrew J. Frank, Christopher DuBois
2021 arXiv   pre-print
Existing methods for open-retrieval question answering in lower resource languages (LRLs) lag significantly behind English.  ...  Assuming a strong English question answering model or database, we compare and analyze methods that pivot through English: to map foreign queries to English and then English answers back to target language  ...  Task: Cross-Lingual Pivots The Open-Retrieval Question Answering (ORQA) task evaluates models' ability to answer information-seeking questions.  ... 
arXiv:2012.14094v2 fatcat:bbvne32bxvhixowjyq5fewpxre

Investigating Post-pretraining Representation Alignment for Cross-Lingual Question Answering [article]

Fahim Faisal, Antonios Anastasopoulos
2021 arXiv   pre-print
Hence, for information-seeking question answering (QA) systems to adequately serve speakers of all languages, they need to operate cross-lingually.  ...  In this work we investigate the capabilities of multilingually pre-trained language models on cross-lingual QA.  ...  We also want to thank Jacob Eisenstein, Manaal Faruqui, and Jon Clark for helpful discussions on question answering and data collection.  ... 
arXiv:2109.12028v1 fatcat:hvox6jcsgrcdhmzf5thqvis7my

A Survey on non-English Question Answering Dataset [article]

Andreas Chandra, Affandy Fahrizain, Ibrahim, Simon Willyanto Laufried
2021 arXiv   pre-print
cross-lingual question-answering datasets.  ...  Research in question answering datasets and models has gained a lot of attention in the research community. Many of them release their own question answering datasets as well as the models.  ...  The paper proposes a new task called cross-lingual open-retrieval question-answering and datasets called XOR-TYDI-QA which are derived from TYD-QA.  ... 
arXiv:2112.13634v1 fatcat:l2dvllft35ga7b3tfjy6tgxcda

Cross-lingual Passage Re-ranking with Alignment Augmented Multilingual BERT

Dongmei Chen, Sheng Zhang, Xin Zhang, Kaijing Yang
2020 IEEE Access  
Two latest works, i.e., LAReQA [16] and XOR QA [17] , focus on cross-lingual question answering and are more closely related to ours.  ...  INTRODUCTION Passage re-ranking is an essential task in many Natural Language Processing (NLP) applications such as passage retrieval for open-domain question answering.  ... 
doi:10.1109/access.2020.3041605 fatcat:ppp72cpcgzhkpbsc6sy2th3yru

Towards Best Practices for Training Multilingual Dense Retrieval Models [article]

Xinyu Zhang, Kelechi Ogueji, Xueguang Ma, Jimmy Lin
2022 arXiv   pre-print
Although recent work with multilingual transformers demonstrates that they exhibit strong cross-lingual generalization capabilities, there remain many open research questions, which we tackle here.  ...  In considering these scenarios, we gain a better understanding of the role of multi-stage fine-tuning, the strength of cross-lingual transfer under various conditions, the usefulness of out-of-language  ...  To our knowledge, this is the only available dataset that meets our needs (i.e., monolingual retrieval); other datasets, for example, XOR-TYDI [3] and CLIRMatrix [28] focus on cross-lingual retrieval  ... 
arXiv:2204.02363v1 fatcat:qeqf3klwi5brjauxsepcgff3gq

SD-QA: Spoken Dialectal Question Answering for the Real World

Fahim Faisal, Sharlina Keshava, Md Mahfuz Ibn Alam, Antonios Anastasopoulos
2021 Findings of the Association for Computational Linguistics: EMNLP 2021   unpublished
Question answering (QA) systems are now available through numerous commercial applications for a wide variety of domains, serving millions of users that interact with them via speech interfaces.  ...  To address this gap, we augment an existing QA dataset to construct a multi-dialect, spoken QA benchmark on five languages (Arabic, Bengali, English, Kiswahili, Korean) with more than 68k audio prompts  ...  We also want to thank Jacob Eisenstein, Manaal Faruqui, and Jon Clark for helpful discussions on question answering and data collection.  ... 
doi:10.18653/v1/2021.findings-emnlp.281 fatcat:fjhzgyc2nbfe5m5hkfuohvvl2m
« Previous Showing results 1 — 15 out of 24 results