no code implementations • Findings (EMNLP) 2021 • Chang Xu, Jun Wang, Francisco Guzmán, Benjamin Rubinstein, Trevor Cohn
NLP models are vulnerable to data poisoning attacks.
no code implementations • ALTA 2021 • Najam Zaidi, Trevor Cohn, Gholamreza Haffari
In this paper, we present a novel semi-autoregressive document generation model capable of revising and editing the generated text.
no code implementations • ACL 2022 • Jun Wang, Benjamin Rubinstein, Trevor Cohn
In this paper we describe a new source of bias prevalent in NMT systems, relating to translations of sentences containing person names.
no code implementations • 14 Dec 2024 • Sukai Huang, Trevor Cohn, Nir Lipovetzky
The capability of Large Language Models (LLMs) to plan remains a topic of debate.
no code implementations • 24 Sep 2024 • Sukai Huang, Nir Lipovetzky, Trevor Cohn
Large Language Models (LLMs) have shown promise in solving natural language-described planning tasks, but their direct use often leads to inconsistent reasoning and hallucination.
no code implementations • 24 Sep 2024 • Sukai Huang, Shu-Wei Liu, Nir Lipovetzky, Trevor Cohn
While Vision-Language Models (VLMs) are increasingly used to generate reward signals for training embodied agents to follow instructions, our research reveals that agents guided by VLM rewards often underperform compared to those employing only intrinsic (exploration-driven) rewards, contradicting expectations set by recent work.
no code implementations • 20 Sep 2024 • Zheng Wei Lim, Nitish Gupta, Honglin Yu, Trevor Cohn
Multilingual large language models (LLMs) are great translators, but this is largely limited to high-resource languages.
no code implementations • 15 Jul 2024 • Jun Wang, Eleftheria Briakou, Hamid Dadkhahi, Rishabh Agarwal, Colin Cherry, Trevor Cohn
A critical component in knowledge distillation is the means of coupling the teacher and student.
no code implementations • 19 May 2024 • Xuanli He, Qiongkai Xu, Jun Wang, Benjamin I. P. Rubinstein, Trevor Cohn
Modern NLP models are often trained on public datasets drawn from diverse sources, rendering them vulnerable to data poisoning attacks.
no code implementations • 30 Apr 2024 • Xuanli He, Jun Wang, Qiongkai Xu, Pasquale Minervini, Pontus Stenetorp, Benjamin I. P. Rubinstein, Trevor Cohn
The implications of backdoor attacks on English-centric large language models (LLMs) have been widely examined - such attacks can be achieved by embedding malicious behaviors during training and activated under specific conditions that trigger malicious outputs.
no code implementations • 3 Apr 2024 • Thinh Hung Truong, Yulia Otmakhova, Karin Verspoor, Trevor Cohn, Timothy Baldwin
In this work, we measure the impact of affixal negation on modern English large language models (LLMs).
no code implementations • 3 Apr 2024 • Jun Wang, Qiongkai Xu, Xuanli He, Benjamin I. P. Rubinstein, Trevor Cohn
Our aim is to bring attention to these vulnerabilities within MNMT systems with the hope of encouraging the community to address security concerns in machine translation, especially in the context of low-resource languages.
1 code implementation • 26 Feb 2024 • Fan Jiang, Tom Drummond, Trevor Cohn
Cross-lingual open domain question answering (CLQA) is a complex problem, comprising cross-lingual retrieval from a multilingual knowledge base, followed by answer generation in the query language.
no code implementations • 20 Feb 2024 • Zheng Wei Lim, Ekaterina Vylomova, Trevor Cohn, Charles Kemp
On one hand, intuition and some prior work suggest that accuracy and fluency should trade off against each other, and that capturing every detail of the source can only be achieved at the cost of fluency.
no code implementations • 19 Dec 2023 • Zheng Wei Lim, Ekaterina Vylomova, Charles Kemp, Trevor Cohn
Human translators linger on some words and phrases more than others, and predicting this variation is a step towards explaining the underlying cognitive processes.
1 code implementation • 27 Nov 2023 • Fan Jiang, Tom Drummond, Trevor Cohn
Although existing neural retrieval models reveal promising results when training data is abundant and the performance keeps improving as training data increases, collecting high-quality annotated data is prohibitively costly.
1 code implementation • 27 Nov 2023 • Fan Jiang, Qiongkai Xu, Tom Drummond, Trevor Cohn
Experimental results demonstrate that our unsupervised $\texttt{ABEL}$ model outperforms both leading supervised and unsupervised retrievers on the BEIR benchmark.
1 code implementation • 3 Nov 2023 • Jinrui Yang, Timothy Baldwin, Trevor Cohn
We present Multi-EuP, a new multilingual benchmark dataset, comprising 22K multi-lingual documents collected from the European Parliament, spanning 24 languages.
1 code implementation • 12 Sep 2023 • Qiongkai Xu, Trevor Cohn, Olga Ohrimenko
Federated Learning allows collaborative training without data sharing in settings where participants do not trust the central server and one another.
1 code implementation • 14 Jun 2023 • Thinh Hung Truong, Timothy Baldwin, Karin Verspoor, Trevor Cohn
Negation has been shown to be a major bottleneck for masked language models, such as BERT.
1 code implementation • 26 May 2023 • Sukai Huang, Nir Lipovetzky, Trevor Cohn
Teaching agents to follow complex written instructions has been an important yet elusive goal.
1 code implementation • 25 May 2023 • Xuanli He, Jun Wang, Benjamin Rubinstein, Trevor Cohn
Backdoor attacks are an insidious security threat against machine learning models.
1 code implementation • 19 May 2023 • Xuanli He, Qiongkai Xu, Jun Wang, Benjamin Rubinstein, Trevor Cohn
Modern NLP models are often trained over large untrusted datasets, raising the potential for a malicious adversary to compromise model behaviour.
1 code implementation • 15 Mar 2023 • Zhuohan Xie, Miao Li, Trevor Cohn, Jey Han Lau
Numerous evaluation metrics have been developed for natural language generation tasks, but their effectiveness in evaluating stories is limited as they are not specifically tailored to assess intricate aspects of storytelling, such as fluency and interestingness.
1 code implementation • 11 Feb 2023 • Xudong Han, Timothy Baldwin, Trevor Cohn
Modern NLP systems exhibit a range of biases, which a growing literature on model debiasing attempts to correct.
no code implementations • 24 Jan 2023 • Zhuohan Xie, Trevor Cohn, Jey Han Lau
To enhance the quality of generated stories, recent story generation models have been investigating the utilization of higher-level attributes like plots or commonsense knowledge.
no code implementations • 15 Nov 2022 • Qin Zhang, Shangsi Chen, Dongkuan Xu, Qingqing Cao, Xiaojun Chen, Trevor Cohn, Meng Fang
Thus, a trade-off between accuracy, memory consumption and processing speed is pursued.
1 code implementation • 17 Oct 2022 • Xudong Han, Aili Shen, Trevor Cohn, Timothy Baldwin, Lea Frermann
Mitigating bias in training on biased datasets is an important open problem.
no code implementations • 11 Oct 2022 • You Guo, Jun Wang, Trevor Cohn
Deep neural networks are vulnerable to adversarial attacks, such as backdoor attacks in which a malicious adversary compromises a model during training such that specific behaviour can be triggered at test time by attaching a specific word or phrase to an input.
1 code implementation • 6 Oct 2022 • Thinh Hung Truong, Yulia Otmakhova, Timothy Baldwin, Trevor Cohn, Jey Han Lau, Karin Verspoor
Negation is poorly captured by current language models, although the extent of this problem is not widely understood.
2 code implementations • sdp (COLING) 2022 • Yulia Otmakhova, Hung Thinh Truong, Timothy Baldwin, Trevor Cohn, Karin Verspoor, Jey Han Lau
In this paper we report on our submission to the Multidocument Summarisation for Literature Review (MSLR) shared task.
1 code implementation • 15 Sep 2022 • Terry Yue Zhuo, Qiongkai Xu, Xuanli He, Trevor Cohn
Round-trip translation could be served as a clever and straightforward technique to alleviate the requirement of the parallel evaluation corpus.
no code implementations • NAACL 2022 • Thinh Hung Truong, Timothy Baldwin, Trevor Cohn, Karin Verspoor
Negation is a common linguistic feature that is crucial in many language understanding tasks, yet it remains a hard problem due to diversity in its expression in different types of text.
1 code implementation • NAACL 2022 • Aili Shen, Xudong Han, Trevor Cohn, Timothy Baldwin, Lea Frermann
Real-world datasets often encode stereotypes and societal biases.
2 code implementations • 4 May 2022 • Xudong Han, Aili Shen, Yitong Li, Lea Frermann, Timothy Baldwin, Trevor Cohn
This paper presents fairlib, an open-source framework for assessing and improving classification fairness.
1 code implementation • 12 Mar 2022 • Xudong Han, Timothy Baldwin, Trevor Cohn
Adversarial training is a common approach for bias mitigation in natural language processing.
1 code implementation • 22 Feb 2022 • Fan Jiang, Trevor Cohn
Moreover, we also explore to utilise higher-order neighbourhood information to encode rich structures in constituent trees.
no code implementations • 16 Feb 2022 • Thinh Hung Truong, Yulia Otmakhova, Rahmad Mahendra, Timothy Baldwin, Jey Han Lau, Trevor Cohn, Lawrence Cavedon, Damiano Spina, Karin Verspoor
This paper describes the submissions of the Natural Language Processing (NLP) team from the Australian Research Council Industrial Transformation Training Centre (ITTC) for Cognitive Computing in Medical Technologies to the TREC 2021 Clinical Trials Track.
no code implementations • ALTA 2021 • Zhuohan Xie, Trevor Cohn, Jey Han Lau
GPT-2 has been frequently adapted in story generation models as it provides powerful generative capability.
1 code implementation • EMNLP 2021 • Jinming Zhao, Philip Arthur, Gholamreza Haffari, Trevor Cohn, Ehsan Shareghi
Most existing simultaneous machine translation (SiMT) systems are trained and evaluated on offline translation corpora.
1 code implementation • NAACL 2022 • Kemal Kurniawan, Lea Frermann, Philip Schulz, Trevor Cohn
Providing technologies to communities or domains where training data is scarce or protected e. g., for privacy reasons, is becoming increasingly important.
no code implementations • 22 Sep 2021 • Aili Shen, Xudong Han, Trevor Cohn, Timothy Baldwin, Lea Frermann
Trained classification models can unintentionally lead to biased representations and predictions, which can reinforce societal preconceptions and stereotypes.
no code implementations • EMNLP 2021 • Shivashankar Subramanian, Xudong Han, Timothy Baldwin, Trevor Cohn, Lea Frermann
Bias is pervasive in NLP models, motivating the development of automatic debiasing techniques.
no code implementations • EMNLP 2021 • Shivashankar Subramanian, Afshin Rahimi, Timothy Baldwin, Trevor Cohn, Lea Frermann
Class imbalance is a common challenge in many NLP tasks, and has clear connections to bias, in that bias in training data often leads to higher accuracy for majority groups at the expense of minority groups.
1 code implementation • CoNLL (EMNLP) 2021 • Chunhua Liu, Trevor Cohn, Lea Frermann
Humans use countless basic, shared facts about the world to efficiently navigate in their environment.
no code implementations • 16 Sep 2021 • Xudong Han, Timothy Baldwin, Trevor Cohn
Group bias in natural language processing tasks manifests as disparities in system error rates across texts authorized by different demographic groups, typically disadvantaging minority groups.
1 code implementation • INLG (ACL) 2021 • Jiuzhou Han, Daniel Beck, Trevor Cohn
Text generation from semantic graphs is traditionally performed with deterministic methods, which generate a unique description given an input graph.
no code implementations • SEMEVAL 2021 • Kemal Kurniawan, Lea Frermann, Philip Schulz, Trevor Cohn
This paper describes PTST, a source-free unsupervised domain adaptation technique for sequence tagging, and its application to the SemEval-2021 Task 10 on time expression recognition.
1 code implementation • Findings (ACL) 2021 • Jun Wang, Chang Xu, Francisco Guzman, Ahmed El-Kishky, Benjamin I. P. Rubinstein, Trevor Cohn
Mistranslated numbers have the potential to cause serious effects, such as financial loss or medical misinformation.
1 code implementation • 12 Jul 2021 • Jun Wang, Chang Xu, Francisco Guzman, Ahmed El-Kishky, Yuqing Tang, Benjamin I. P. Rubinstein, Trevor Cohn
Neural machine translation systems are known to be vulnerable to adversarial test inputs, however, as we show in this paper, these systems are also vulnerable to training attacks.
1 code implementation • NAACL 2021 • Fan Jiang, Trevor Cohn
External syntactic and semantic information has been largely ignored by existing neural coreference resolution models.
1 code implementation • NAACL 2021 • Shima Khanehzar, Trevor Cohn, Gosia Mikolajczak, Andrew Turpin, Lea Frermann
Understanding how news media frame political issues is important due to its impact on public attitudes, yet hard to automate.
1 code implementation • EACL 2021 • Kemal Kurniawan, Lea Frermann, Philip Schulz, Trevor Cohn
Cross-lingual transfer is a leading technique for parsing low-resource languages in the absence of explicit supervision.
1 code implementation • EACL 2021 • Xudong Han, Timothy Baldwin, Trevor Cohn
Adversarial learning can learn fairer and less biased models of language than standard methods.
no code implementations • 2 Nov 2020 • Chang Xu, Jun Wang, Yuqing Tang, Francisco Guzman, Benjamin I. P. Rubinstein, Trevor Cohn
In this paper, we show that targeted attacks on black-box NMT systems are feasible, based on poisoning a small fraction of their parallel training data.
1 code implementation • ACL 2020 • Nitika Mathur, Timothy Baldwin, Trevor Cohn
Automatic metrics are fundamental for the development and evaluation of machine translation systems.
no code implementations • ICLR 2020 • Najam Zaidi, Trevor Cohn, Gholamreza Haffari
Decoding in autoregressive models (ARMs) consists of searching for a high scoring output sequence under the trained model.
no code implementations • EACL 2021 • Philip Arthur, Trevor Cohn, Gholamreza Haffari
We present a novel approach to efficiently learn a simultaneous translation model with coupled programmer-interpreter policies.
no code implementations • WS 2019 • Daniel Beck, Trevor Cohn, Gholamreza Haffari
Speech translation systems usually follow a pipeline approach, using word lattices as an intermediate representation.
no code implementations • WS 2019 • Dalin Wang, Daniel Beck, Trevor Cohn
Scene graphs represent semantic information in images, which can help image captioning system to produce more descriptive outputs versus using only the image as context.
1 code implementation • IJCNLP 2019 • Xudong Han, Philip Schulz, Trevor Cohn
In addition, we present a model that operates in the HSV color space.
1 code implementation • IJCNLP 2019 • Shivashankar Subramanian, Trevor Cohn, Timothy Baldwin
Many pledges are made in the course of an election campaign, forming important corpora for political analysis of campaign strategy and governmental accountability.
1 code implementation • WS 2019 • Zenan Zhai, Dat Quoc Nguyen, Saber A. Akhondi, Camilo Thorne, Christian Druckenbrodt, Trevor Cohn, Michelle Gregory, Karin Verspoor
In this paper, we explore the NER performance of a BiLSTM-CRF model utilising pre-trained word embeddings, character-level word representations and contextualized ELMo word representations for chemical patents.
1 code implementation • ACL 2019 • Nitika Mathur, Timothy Baldwin, Trevor Cohn
Accurate, automatic evaluation of machine translation is critical for system tuning, and evaluating progress in the field.
no code implementations • ACL 2019 • Yitong Li, Timothy Baldwin, Trevor Cohn
Supervised models of NLP rely on large collections of text which closely resemble the intended testing setting.
1 code implementation • SEMEVAL 2019 • Shivashankar Subramanian, Trevor Cohn, Timothy Baldwin
We study pragmatics in political campaign text, through analysis of speech acts and the target of each utterance.
no code implementations • NAACL 2019 • Ekaterina Vylomova, Ryan Cotterell, Timothy Baldwin, Trevor Cohn, Jason Eisner
Critical to natural language generation is the production of correctly inflected text.
no code implementations • ALTA 2019 • Zhuohan Xie, Jey Han Lau, Trevor Cohn
In this paper, we adapt Deep-speare, a joint neural network model for English sonnets, to Chinese poetry.
1 code implementation • 1 Mar 2019 • Steven Spratley, Daniel Beck, Trevor Cohn
Within Music Information Retrieval (MIR), prominent tasks -- including pitch-tracking, source-separation, super-resolution, and synthesis -- typically call for specialised methods, despite their similarities.
no code implementations • 24 Feb 2019 • Yuan Li, Benjamin I. P. Rubinstein, Trevor Cohn
As we show, datasets produced by crowd-sourcing are often not of this type: the data is highly redundantly annotated ($\ge 5$ annotations per instance), and the vast majority of workers produce high quality outputs.
1 code implementation • ACL 2019 • Afshin Rahimi, Yuan Li, Trevor Cohn
In cross-lingual transfer, NLP models over one or more source languages are applied to a low-resource target language.
no code implementations • ALTA 2018 • Nitika Mathur, Timothy Baldwin, Trevor Cohn
In this paper we show that the quality control mechanism is overly conservative, which increases the time and expense of the evaluation.
no code implementations • ALTA 2018 • Cong Duy Vu Hoang, Gholamreza Haffari, Trevor Cohn
In this work, we investigate whether side information is helpful in neural machine translation (NMT).
no code implementations • WS 2018 • Taro Miyazaki, Afshin Rahimi, Trevor Cohn, Timothy Baldwin
Automatic geolocation of microblog posts from their text content is particularly difficult because many location-indicative terms are rare terms, notably entity names such as locations, people or local organisations.
1 code implementation • EMNLP 2018 • Minghao Wu, Fei Liu, Trevor Cohn
Conventional wisdom is that hand-crafted features are redundant for deep learning models, as they already learn adequate representations of text automatically from corpora.
Ranked #45 on
Named Entity Recognition (NER)
on CoNLL 2003 (English)
1 code implementation • ACL 2018 • Jey Han Lau, Trevor Cohn, Timothy Baldwin, Julian Brooke, Adam Hammond
In this paper, we propose a joint architecture that captures language, rhyme and meter for sonnet modelling.
no code implementations • WS 2018 • Vu Cong Duy Hoang, Philipp Koehn, Gholamreza Haffari, Trevor Cohn
We present iterative back-translation, a method for generating increasingly better synthetic parallel data from monolingual data to train neural machine translation systems.
2 code implementations • ACL 2018 • Daniel Beck, Gholamreza Haffari, Trevor Cohn
Many NLP applications can be framed as a graph-to-sequence learning problem.
1 code implementation • ACL 2018 • Philip Schulz, Wilker Aziz, Trevor Cohn
The process of translation is ambiguous, in that there are typically many valid trans- lations for a given sentence.
1 code implementation • ACL 2018 • Shivashankar Subramanian, Timothy Baldwin, Trevor Cohn
Online petitions are a cost-effective way for citizens to collectively engage with policy-makers in a democracy.
1 code implementation • NAACL 2018 • Yitong Li, Timothy Baldwin, Trevor Cohn
Most real world language problems require learning from heterogenous corpora, raising the problem of learning robust models which generalise well to both similar (in domain) and dissimilar (out of domain) instances to those seen in training.
1 code implementation • ACL 2018 • Fei Liu, Trevor Cohn, Timothy Baldwin
Story comprehension requires a deep semantic understanding of the narrative, making it a challenging task.
3 code implementations • ACL 2018 • Yitong Li, Timothy Baldwin, Trevor Cohn
Written text often provides sufficient clues to identify the author, their gender, age, and other important attributes.
no code implementations • NAACL 2018 • Shivashankar Subramanian, Trevor Cohn, Timothy Baldwin
Election manifestos document the intentions, motives, and views of political parties.
1 code implementation • NAACL 2018 • Fei Liu, Trevor Cohn, Timothy Baldwin
While neural networks have been shown to achieve impressive results for sentence-level sentiment analysis, targeted aspect-based sentiment analysis (TABSA) --- extraction of fine-grained opinion polarity w. r. t.
Ranked #3 on
Aspect-Based Sentiment Analysis (ABSA)
on Sentihood
Aspect-Based Sentiment Analysis
Aspect-Based Sentiment Analysis (ABSA)
+2
1 code implementation • ACL 2018 • Afshin Rahimi, Trevor Cohn, Timothy Baldwin
Social media user geolocation is vital to many applications such as event detection.
no code implementations • 6 Dec 2017 • Arkaitz Zubiaga, Elena Kochkina, Maria Liakata, Rob Procter, Michal Lukasik, Kalina Bontcheva, Trevor Cohn, Isabelle Augenstein
We show that sequential classifiers that exploit the use of discourse properties in social media conversations while using only local features, outperform non-sequential classifiers.
no code implementations • IJCNLP 2017 • Daniel Beck, Trevor Cohn
Non-contiguous word sequences are widely known to be important in modelling natural language.
1 code implementation • IJCNLP 2017 • Jey Han Lau, Lianhua Chi, Khoi-Nguyen Tran, Trevor Cohn
We propose an end-to-end neural network to predict the geolocation of a tweet.
1 code implementation • IJCNLP 2017 • Fei Liu, Timothy Baldwin, Trevor Cohn
Despite successful applications across a broad range of NLP tasks, conditional random fields ("CRFs"), in particular the linear-chain variant, are only able to model local features.
no code implementations • EMNLP 2017 • Cong Duy Vu Hoang, Gholamreza Haffari, Trevor Cohn
We propose a novel decoding approach for neural machine translation (NMT) based on continuous optimisation.
no code implementations • WS 2017 • Yitong Li, Trevor Cohn, Timothy Baldwin
This paper describes our submission to the sentiment analysis sub-task of {``}Build It, Break It: The Language Edition (BIBI){''}, on both the builder and breaker sides.
no code implementations • EMNLP 2017 • Nitika Mathur, Timothy Baldwin, Trevor Cohn
Manual data annotation is a vital component of NLP research.
1 code implementation • EMNLP 2017 • Afshin Rahimi, Timothy Baldwin, Trevor Cohn
We propose a method for embedding two-dimensional locations in a continuous vector space using a neural network-based model incorporating mixtures of Gaussian distributions, presenting two model variants for text-based geolocation and lexical dialectology.
1 code implementation • EMNLP 2017 • Meng Fang, Yuan Li, Trevor Cohn
Active learning aims to select a small subset of data for annotation such that a classifier learned on the data is highly accurate.
1 code implementation • ACL 2017 • Meng Fang, Trevor Cohn
Cross-lingual model transfer is a compelling and popular method for predicting annotations in a low-resource language, whereby parallel corpora provide a bridge to a high-resource language and its associated annotated corpora.
1 code implementation • ACL 2017 • Jey Han Lau, Timothy Baldwin, Trevor Cohn
Language models are typically applied at the sentence level, without access to the broader document context.
no code implementations • ACL 2017 • Afshin Rahimi, Trevor Cohn, Timothy Baldwin
We propose a simple yet effective text- based user geolocation model based on a neural network with one hidden layer, which achieves state of the art performance over three Twitter benchmark geolocation datasets, in addition to producing word and phrase embeddings in the hidden layer that we show to be useful for detecting dialectal terms.
1 code implementation • EACL 2017 • Yitong Li, Trevor Cohn, Timothy Baldwin
Deep neural networks have achieved remarkable results across many language processing tasks, however they have been shown to be susceptible to overfitting and highly sensitive to noise, including adversarial attacks.
no code implementations • EACL 2017 • Oliver Adams, Adam Makarucha, Graham Neubig, Steven Bird, Trevor Cohn
We investigate the use of such lexicons to improve language models when textual training data is limited to as few as a thousand sentences.
no code implementations • EACL 2017 • Long Duong, Hiroshi Kanayama, Tengfei Ma, Steven Bird, Trevor Cohn
Crosslingual word embeddings represent lexical items from different languages using the same vector space, enabling crosslingual transfer.
no code implementations • WS 2017 • Ying Xu, Jey Han Lau, Timothy Baldwin, Trevor Cohn
With this decoupled architecture, we decrease the number of parameters in the decoder substantially, and shorten its training time.
1 code implementation • EACL 2017 • Ekaterina Vylomova, Ryan Cotterell, Timothy Baldwin, Trevor Cohn
Derivational morphology is a fundamental and complex characteristic of language.
4 code implementations • 15 Jan 2017 • Graham Neubig, Chris Dyer, Yoav Goldberg, Austin Matthews, Waleed Ammar, Antonios Anastasopoulos, Miguel Ballesteros, David Chiang, Daniel Clothiaux, Trevor Cohn, Kevin Duh, Manaal Faruqui, Cynthia Gan, Dan Garrette, Yangfeng Ji, Lingpeng Kong, Adhiguna Kuncoro, Gaurav Kumar, Chaitanya Malaviya, Paul Michel, Yusuke Oda, Matthew Richardson, Naomi Saphra, Swabha Swayamdipta, Pengcheng Yin
In the static declaration strategy that is used in toolkits like Theano, CNTK, and TensorFlow, the user first defines a computation graph (a symbolic representation of the computation), and then examples are fed into an engine that executes this computation and computes its derivatives.
no code implementations • 11 Jan 2017 • Cong Duy Vu Hoang, Gholamreza Haffari, Trevor Cohn
We propose a novel decoding approach for neural machine translation (NMT) based on continuous optimisation.
no code implementations • COLING 2016 • Matthias Petri, Trevor Cohn
Succinct data structures involve the use of novel data structures, compression technologies, and other mechanisms to allow data to be stored in extremely small memory or disk footprints, while still allowing for efficient access to the underlying data.
1 code implementation • EMNLP 2016 • Yitong Li, Trevor Cohn, Timothy Baldwin
Deep neural networks have achieved remarkable results across many language processing tasks, however these methods are highly sensitive to noise and adversarial attacks.
no code implementations • 7 Sep 2016 • Michal Lukasik, Kalina Bontcheva, Trevor Cohn, Arkaitz Zubiaga, Maria Liakata, Rob Procter
Social media tend to be rife with rumours while new reports are released piecemeal during breaking news.
1 code implementation • TACL 2016 • Ehsan Shareghi, Matthias Petri, Gholamreza Haffari, Trevor Cohn
Efficient methods for storing and querying are critical for scaling high-order n-gram language models to large corpora.
no code implementations • CONLL 2016 • Meng Fang, Trevor Cohn
Cross lingual projection of linguistic annotation suffers from many sources of bias and noise, leading to unreliable annotations that cannot be used directly.
1 code implementation • EMNLP 2016 • Long Duong, Hiroshi Kanayama, Tengfei Ma, Steven Bird, Trevor Cohn
Crosslingual word embeddings represent lexical items from different languages in the same vector space, enabling transfer of NLP tools.
Bilingual Lexicon Induction
Cross-Lingual Document Classification
+4
no code implementations • CONLL 2016 • Daniel Beck, Lucia Specia, Trevor Cohn
Machine Translation Quality Estimation is a notoriously difficult task, which lessens its usefulness in real-world translation environments.
no code implementations • WS 2017 • Ekaterina Vylomova, Trevor Cohn, Xuanli He, Gholamreza Haffari
Dealing with the complex word forms in morphologically rich languages is an open problem in language processing, and is particularly important in translation.
no code implementations • LREC 2016 • Daniel Preo{\c{t}}iuc-Pietro, P. K. Srijith, Mark Hepple, Trevor Cohn
Streaming media provides a number of unique challenges for computational linguistics.
no code implementations • NAACL 2016 • Trevor Cohn, Cong Duy Vu Hoang, Ekaterina Vymolova, Kaisheng Yao, Chris Dyer, Gholamreza Haffari
Neural encoder-decoder models of machine translation have achieved impressive results, rivalling traditional translation models.
1 code implementation • 12 Nov 2015 • Yangfeng Ji, Trevor Cohn, Lingpeng Kong, Chris Dyer, Jacob Eisenstein
Text documents are structured on multiple levels of detail: individual words are related by syntax, but larger units of text are related by discourse structure.
1 code implementation • ACL 2016 • Ekaterina Vylomova, Laura Rimell, Trevor Cohn, Timothy Baldwin
Recent work on word embeddings has shown that simple vector subtraction over pre-trained embeddings is surprisingly effective at capturing different lexical relations, despite lacking explicit supervision.
no code implementations • 16 Aug 2015 • Kaisheng Yao, Trevor Cohn, Katerina Vylomova, Kevin Duh, Chris Dyer
This gate is a function of the lower layer memory cell, the input to and the past memory cell of this layer.
no code implementations • TACL 2015 • Daniel Beck, Trevor Cohn, Christian Hardmeier, Lucia Specia
Structural kernels are a flexible learning paradigm that has been widely used in Natural Language Processing.
no code implementations • IJCNLP 2015 • Afshin Rahimi, Trevor Cohn, Timothy Baldwin
We propose a label propagation approach to geolocation prediction based on Modified Adsorption, with two enhancements:(1) the removal of "celebrity" nodes to increase location homophily and boost tractability, and (2) he incorporation of text-based geolocation priors for test users.
no code implementations • HLT 2015 • Afshin Rahimi, Duy Vu, Trevor Cohn, Timothy Baldwin
Research on automatically geolocating social media users has conventionally been based on the text content of posts from a given user or the social network of the user, with very little crossover between the two, and no bench-marking of the two approaches over compara- ble datasets.
no code implementations • EMNLP 2015 • Michal Lukasik, Trevor Cohn, Kalina Bontcheva
Social media is a rich source of rumours and corresponding community reactions.
no code implementations • 9 Mar 2015 • Ehsan Shareghi, Gholamreza Haffari, Trevor Cohn, Ann Nicholson
Linguistic structures exhibit a rich array of global phenomena, however commonly used Markov models are unable to adequately describe these phenomena due to their strong locality assumptions.
no code implementations • 1 Oct 2010 • Phil Blunsom, Trevor Cohn
Inducing a grammar directly from text is one of the oldest and most challenging tasks in Computational Linguistics.
Ranked #4 on
Unsupervised Dependency Parsing
on Penn Treebank
Dependency Grammar Induction
Unsupervised Dependency Parsing
no code implementations • NeurIPS 2008 • Phil Blunsom, Trevor Cohn, Miles Osborne
We present a novel method for inducing synchronous context free grammars (SCFGs) from a corpus of parallel string pairs.