no code implementations • EMNLP 2020 • Ramy Eskander, Smaranda Muresan, Michael Collins
Our approach innovates in three ways: 1) a robust approach of selecting training instances via cross-lingual annotation projection that exploits best practices of unsupervised type and token constraints, word-alignment confidence and density of projected POS, 2) a Bi-LSTM architecture that uses contextualized word embeddings, affix embeddings and hierarchical Brown clusters, and 3) an evaluation on 12 diverse languages in terms of language family and morphological typology.
no code implementations • NAACL (DeeLIO) 2021 • Vidhisha Balachandran, Bhuwan Dhingra, Haitian Sun, Michael Collins, William Cohen
We create a subset of the NQ data, Factual Questions (FQ), where the questions have evidence in the KB in the form of paths that link question entities to answer entities but still must be answered using text, to facilitate further research into KB integration methods.
no code implementations • 27 Dec 2024 • Ananth Balashankar, Ziteng Sun, Jonathan Berant, Jacob Eisenstein, Michael Collins, Adrian Hutter, Jong Lee, Chirag Nagpal, Flavien Prost, Aradhana Sinha, Ananda Theertha Suresh, Ahmad Beirami
We prove that for any inference-time decoding procedure, the optimal aligned policy is the solution to the standard RLHF problem with a transformation of the reward.
no code implementations • 31 May 2024 • Bernd Bohnet, Kevin Swersky, Rosanne Liu, Pranjal Awasthi, Azade Nova, Javier Snaider, Hanie Sedghi, Aaron T Parisi, Michael Collins, Angeliki Lazaridou, Orhan Firat, Noah Fiedel
We explore the use of long-context capabilities in large language models to create synthetic reading comprehension data from entire books.
no code implementations • 1 Feb 2024 • Alon Jacovi, Yonatan Bitton, Bernd Bohnet, Jonathan Herzig, Or Honovich, Michael Tseng, Michael Collins, Roee Aharoni, Mor Geva
REVEAL includes comprehensive labels for the relevance, attribution to evidence passages, and logical correctness of each reasoning step in a language model's answer, across a variety of datasets and state-of-the-art language models.
no code implementations • 25 Oct 2023 • Sidharth Mudgal, Jong Lee, Harish Ganapathy, Yaguang Li, Tao Wang, Yanping Huang, Zhifeng Chen, Heng-Tze Cheng, Michael Collins, Trevor Strohman, Jilin Chen, Alex Beutel, Ahmad Beirami
KL-regularized reinforcement learning (RL) is a popular alignment framework to control the language model responses towards high reward outcomes.
no code implementations • 22 Jan 2023 • Christopher Mohri, Daniel Andor, Eunsol Choi, Michael Collins
We study the problem of classification with a reject option for a fixed predictor, applicable in natural language processing.
1 code implementation • 15 Dec 2022 • Bernd Bohnet, Vinh Q. Tran, Pat Verga, Roee Aharoni, Daniel Andor, Livio Baldini Soares, Massimiliano Ciaramita, Jacob Eisenstein, Kuzman Ganchev, Jonathan Herzig, Kai Hui, Tom Kwiatkowski, Ji Ma, Jianmo Ni, Lierni Sestorain Saralegui, Tal Schuster, William W. Cohen, Michael Collins, Dipanjan Das, Donald Metzler, Slav Petrov, Kellie Webster
We take human annotations as a gold standard and show that a correlated automatic metric is suitable for development.
1 code implementation • 22 Nov 2022 • Bernd Bohnet, Chris Alberti, Michael Collins
We obtain state-of-the-art accuracy on the CoNLL-2012 datasets with 83. 3 F1-score for English (a 2. 3 higher F1-score than previous work (Dobrovolskii, 2021)) using only CoNLL data for training, 68. 5 F1-score for Arabic (+4. 1 higher than previous work) and 74. 3 F1-score for Chinese (+5. 3).
Ranked #2 on
Coreference Resolution
on OntoNotes
no code implementations • 16 Nov 2022 • Chris Alberti, Kuzman Ganchev, Michael Collins, Sebastian Gehrmann, Ciprian Chelba
Compared to a baseline that generates text using greedy search, we demonstrate two techniques that improve the fluency and semantic accuracy of the generated text: The first technique samples multiple candidate text sequences from which the semantic parser chooses.
no code implementations • 31 Oct 2022 • Reinald Kim Amplayo, Kellie Webster, Michael Collins, Dipanjan Das, Shashi Narayan
Large language models (LLMs) have been shown to perform well in answering questions and in producing long-form texts, both in few-shot closed-book settings.
1 code implementation • 17 Oct 2022 • Thomas Effland, Michael Collins
We present Expected Statistic Regularization (ESR), a novel regularization technique that utilizes low-order multi-task structural statistics to shape model distributions for semi-supervised learning on low-resource datasets.
no code implementations • 5 Oct 2022 • Jacob Eisenstein, Daniel Andor, Bernd Bohnet, Michael Collins, David Mimno
But what sorts of rationales are useful and how can we train systems to produce them?
1 code implementation • ACL 2022 • Shashi Narayan, Gonçalo Simões, Yao Zhao, Joshua Maynez, Dipanjan Das, Michael Collins, Mirella Lapata
We propose Composition Sampling, a simple but effective method to generate diverse outputs for conditional generation of higher quality compared to previous stochastic decoding strategies.
1 code implementation • 23 Dec 2021 • Hannah Rashkin, Vitaly Nikolaev, Matthew Lamm, Lora Aroyo, Michael Collins, Dipanjan Das, Slav Petrov, Gaurav Singh Tomar, Iulia Turc, David Reitter
With recent improvements in natural language generation (NLG) models for various applications, it has become imperative to have the means to identify and evaluate whether NLG output is only sharing verifiable information about the external world.
1 code implementation • 16 Aug 2021 • Thomas Effland, Michael Collins
We study learning named entity recognizers in the presence of missing entity annotations.
no code implementations • 20 May 2021 • Emily Sandford, David Kipping, Michael Collins
A planetary system consists of a host star and one or more planets, arranged into a particular configuration.
no code implementations • 9 Feb 2021 • Eunsol Choi, Jennimaria Palomaki, Matthew Lamm, Tom Kwiatkowski, Dipanjan Das, Michael Collins
Models for question answering, dialogue agents, and summarization often interpret the meaning of a sentence in a rich context and use that meaning in a new context.
no code implementations • 1 Jan 2021 • Sewon Min, Jordan Boyd-Graber, Chris Alberti, Danqi Chen, Eunsol Choi, Michael Collins, Kelvin Guu, Hannaneh Hajishirzi, Kenton Lee, Jennimaria Palomaki, Colin Raffel, Adam Roberts, Tom Kwiatkowski, Patrick Lewis, Yuxiang Wu, Heinrich Küttler, Linqing Liu, Pasquale Minervini, Pontus Stenetorp, Sebastian Riedel, Sohee Yang, Minjoon Seo, Gautier Izacard, Fabio Petroni, Lucas Hosseini, Nicola De Cao, Edouard Grave, Ikuya Yamada, Sonse Shimaoka, Masatoshi Suzuki, Shumpei Miyawaki, Shun Sato, Ryo Takahashi, Jun Suzuki, Martin Fajcik, Martin Docekal, Karel Ondrej, Pavel Smrz, Hao Cheng, Yelong Shen, Xiaodong Liu, Pengcheng He, Weizhu Chen, Jianfeng Gao, Barlas Oguz, Xilun Chen, Vladimir Karpukhin, Stan Peshterliev, Dmytro Okhonko, Michael Schlichtkrull, Sonal Gupta, Yashar Mehdad, Wen-tau Yih
We review the EfficientQA competition from NeurIPS 2020.
1 code implementation • 1 Dec 2020 • Danish Pruthi, Rachit Bansal, Bhuwan Dhingra, Livio Baldini Soares, Michael Collins, Zachary C. Lipton, Graham Neubig, William W. Cohen
While many methods purport to explain predictions by highlighting salient features, what aims these explanations serve and how they ought to be evaluated often go unstated.
1 code implementation • 8 Sep 2020 • Matthew Lamm, Jennimaria Palomaki, Chris Alberti, Daniel Andor, Eunsol Choi, Livio Baldini Soares, Michael Collins
A question answering system that in addition to providing an answer provides an explanation of the reasoning that leads to that answer has potential advantages in terms of debuggability, extensibility and trust.
1 code implementation • 1 May 2020 • Yi Luan, Jacob Eisenstein, Kristina Toutanova, Michael Collins
Dual encoders perform retrieval by encoding documents and queries into dense lowdimensional vectors, scoring each document by its inner product with the query.
2 code implementations • TACL 2020 • Jonathan H. Clark, Eunsol Choi, Michael Collins, Dan Garrette, Tom Kwiatkowski, Vitaly Nikolaev, Jennimaria Palomaki
Confidently making progress on multilingual modeling requires challenging, trustworthy evaluations.
1 code implementation • IJCNLP 2019 • Chris Alberti, Jeffrey Ling, Michael Collins, David Reitter
To advance models of multimodal context, we introduce a simple yet powerful neural architecture for data that combines vision and natural language.
4 code implementations • ACL 2019 • Chris Alberti, Daniel Andor, Emily Pitler, Jacob Devlin, Michael Collins
We introduce a novel method of generating synthetic question answering corpora by combining models of question generation and answer extraction, and by filtering the results to ensure roundtrip consistency.
1 code implementation • Transactions of the Association of Computational Linguistics 2019 • Tom Kwiatkowski, Jennimaria Palomaki, Olivia Redfield, Michael Collins, Ankur Parikh, Chris Alberti, Danielle Epstein, Illia Polosukhin, Jacob Devlin, Kenton Lee, Kristina Toutanova, Llion Jones, Matthew Kelcey, Ming-Wei Chang, Andrew M. Dai, Jakob Uszkoreit, Quoc Le, Slav Petrov
The public release consists of 307, 373 training examples with single annotations, 7, 830 examples with 5-way annotations for development data, and a further 7, 842 examples 5-way annotated sequestered as test data.
Ranked #7 on
Question Answering
on Natural Questions (long)
1 code implementation • NAACL 2019 • Christopher Clark, Kenton Lee, Ming-Wei Chang, Tom Kwiatkowski, Michael Collins, Kristina Toutanova
In this paper we study yes/no questions that are naturally occurring --- meaning that they are generated in unprompted and unconstrained settings.
Ranked #27 on
Question Answering
on BoolQ
no code implementations • NAACL 2019 • Mohammad Sadegh Rasooli, Michael Collins
We describe a cross-lingual transfer method for dependency parsing that takes into account the problem of word order differences between source and target languages.
3 code implementations • 24 Jan 2019 • Chris Alberti, Kenton Lee, Michael Collins
This technical note describes a new baseline for the Natural Questions.
Ranked #6 on
Question Answering
on Natural Questions (long)
no code implementations • 5 Nov 2018 • Hao Cheng, Ming-Wei Chang, Kenton Lee, Ankur Parikh, Michael Collins, Kristina Toutanova
We study approaches to improve fine-grained short answer Question Answering models by integrating coarse-grained data annotated for paragraph-level relevance and show that coarsely annotated data can bring significant performance gains.
no code implementations • EMNLP 2018 • Zhuang Ma, Michael Collins
Noise Contrastive Estimation (NCE) is a powerful parameter estimation method for log-linear models, which avoids calculation of the partition function or its derivatives at each training step, a computationally demanding step in many cases.
Ranked #14 on
Question Answering
on WikiQA
no code implementations • EMNLP 2017 • Yin-Wen Chang, Michael Collins
The algorithm produces a translation by processing the source-language sentence in strictly left-to-right order, differing from commonly used approaches that build the target-language sentence in left-to-right order.
no code implementations • 15 Mar 2017 • Chris Alberti, Daniel Andor, Ivan Bogatyy, Michael Collins, Dan Gillick, Lingpeng Kong, Terry Koo, Ji Ma, Mark Omernick, Slav Petrov, Chayut Thanapirom, Zora Tung, David Weiss
We describe a baseline dependency parsing system for the CoNLL2017 Shared Task.
no code implementations • 13 Jan 2017 • Avner May, Alireza Bagheri Garakani, Zhiyun Lu, Dong Guo, Kuan Liu, Aurélien Bellet, Linxi Fan, Michael Collins, Daniel Hsu, Brian Kingsbury, Michael Picheny, Fei Sha
First, in order to reduce the number of random features required by kernel models, we propose a simple but effective method for feature selection.
no code implementations • TACL 2017 • Yin-Wen Chang, Michael Collins
Decoding of phrase-based translation models in the general case is known to be NP-complete, by a reduction from the traveling salesman problem (Knight, 1999).
1 code implementation • TACL 2017 • Mohammad Sadegh Rasooli, Michael Collins
We describe a simple but effective method for cross-lingual syntactic transfer of dependency parsers, in the scenario where a large amount of translation data is not available.
1 code implementation • ACL 2016 • Daniel Andor, Chris Alberti, David Weiss, Aliaksei Severyn, Alessandro Presta, Kuzman Ganchev, Slav Petrov, Michael Collins
Our model is a simple feed-forward neural network that operates on a task-specific transition system, yet achieves comparable or better accuracies than recurrent models.
Ranked #16 on
Dependency Parsing
on Penn Treebank
no code implementations • 18 Mar 2016 • Zhiyun Lu, Dong Guo, Alireza Bagheri Garakani, Kuan Liu, Avner May, Aurelien Bellet, Linxi Fan, Michael Collins, Brian Kingsbury, Michael Picheny, Fei Sha
We study large-scale kernel methods for acoustic modeling and compare to DNNs on performance metrics related to both acoustic modeling and recognition.
1 code implementation • TACL 2016 • Siva Reddy, Oscar T{\"a}ckstr{\"o}m, Michael Collins, Tom Kwiatkowski, Dipanjan Das, Mark Steedman, Mirella Lapata
In contrast{---}partly due to the lack of a strong type system{---}dependency structures are easy to annotate and have become a widely used form of syntactic analysis for many languages.
1 code implementation • TACL 2016 • Karl Stratos, Michael Collins, Daniel Hsu
We tackle unsupervised part-of-speech (POS) tagging by learning hidden Markov models (HMMs) that are particularly well-suited for the problem.
no code implementations • IJCNLP 2015 • David Weiss, Chris Alberti, Michael Collins, Slav Petrov
We present structured perceptron training for neural network transition-based dependency parsing.
Ranked #17 on
Dependency Parsing
on Penn Treebank
no code implementations • EACL 2014 • Arvind Neelakantan, Michael Collins
This paper describes an approach for automatic construction of dictionaries for Named Entity Recognition (NER) using large amounts of unlabeled data and a few seed examples.
no code implementations • 14 Nov 2014 • Zhiyun Lu, Avner May, Kuan Liu, Alireza Bagheri Garakani, Dong Guo, Aurélien Bellet, Linxi Fan, Michael Collins, Brian Kingsbury, Michael Picheny, Fei Sha
The computational complexity of kernel methods has often been a major barrier for applying them to large-scale learning problems.
Automatic Speech Recognition
Automatic Speech Recognition (ASR)
+2
no code implementations • 23 Jan 2014 • Alexander M. Rush, Michael Collins
Dual decomposition, and more generally Lagrangian relaxation, is a classical method for combinatorial optimization; it has recently been applied to several inference problems in natural language processing (NLP).
no code implementations • NeurIPS 2012 • Michael Collins, Shay B. Cohen
We describe an approach to speed-up inference with latent variable PCFGs, which have been shown to be highly effective for natural language parsing.
no code implementations • NeurIPS 2009 • Natasha Singh-Miller, Michael Collins
We consider the problem of using nearest neighbor methods to provide a conditional probability estimate, P(y|a), when the number of labels y is large and the labels share some underlying structure.