Titles
stringlengths 6
220
| Abstracts
stringlengths 37
3.26k
| Years
int64 1.99k
2.02k
| Categories
stringclasses 1
value |
|---|---|---|---|
Revisiting Pre-Trained Models for Chinese Natural Language Processing | Bidirectional Encoder Representations from Transformers (BERT) has shown
marvelous improvements across various NLP tasks, and consecutive variants have
been proposed to further improve the performance of the pre-trained language
models. In this paper, we target on revisiting Chinese pre-trained language
models to examine their effectiveness in a non-English language and release the
Chinese pre-trained language model series to the community. We also propose a
simple but effective model called MacBERT, which improves upon RoBERTa in
several ways, especially the masking strategy that adopts MLM as correction
(Mac). We carried out extensive experiments on eight Chinese NLP tasks to
revisit the existing pre-trained language models as well as the proposed
MacBERT. Experimental results show that MacBERT could achieve state-of-the-art
performances on many NLP tasks, and we also ablate details with several
findings that may help future research. Resources available:
https://github.com/ymcui/MacBERT
| 2,020 | Computation and Language |
Span-based Localizing Network for Natural Language Video Localization | Given an untrimmed video and a text query, natural language video
localization (NLVL) is to locate a matching span from the video that
semantically corresponds to the query. Existing solutions formulate NLVL either
as a ranking task and apply multimodal matching architecture, or as a
regression task to directly regress the target video span. In this work, we
address NLVL task with a span-based QA approach by treating the input video as
text passage. We propose a video span localizing network (VSLNet), on top of
the standard span-based QA framework, to address NLVL. The proposed VSLNet
tackles the differences between NLVL and span-based QA through a simple yet
effective query-guided highlighting (QGH) strategy. The QGH guides VSLNet to
search for matching video span within a highlighted region. Through extensive
experiments on three benchmark datasets, we show that the proposed VSLNet
outperforms the state-of-the-art methods; and adopting span-based QA framework
is a promising direction to solve NLVL.
| 2,020 | Computation and Language |
Revisiting Round-Trip Translation for Quality Estimation | Quality estimation (QE) is the task of automatically evaluating the quality
of translations without human-translated references. Calculating BLEU between
the input sentence and round-trip translation (RTT) was once considered as a
metric for QE, however, it was found to be a poor predictor of translation
quality. Recently, various pre-trained language models have made breakthroughs
in NLP tasks by providing semantically meaningful word and sentence embeddings.
In this paper, we employ semantic embeddings to RTT-based QE. Our method
achieves the highest correlations with human judgments, compared to previous
WMT 2019 quality estimation metric task submissions. While backward translation
models can be a drawback when using RTT, we observe that with semantic-level
metrics, RTT-based QE is robust to the choice of the backward translation
system. Additionally, the proposed method shows consistent performance for both
SMT and NMT forward translation systems, implying the method does not penalize
a certain type of model.
| 2,020 | Computation and Language |
Evaluating Transformer-Based Multilingual Text Classification | As NLP tools become ubiquitous in today's technological landscape, they are
increasingly applied to languages with a variety of typological structures.
However, NLP research does not focus primarily on typological differences in
its analysis of state-of-the-art language models. As a result, NLP tools
perform unequally across languages with different syntactic and morphological
structures. Through a detailed discussion of word order typology, morphological
typology, and comparative linguistics, we identify which variables most affect
language modeling efficacy; in addition, we calculate word order and
morphological similarity indices to aid our empirical study. We then use this
background to support our analysis of an experiment we conduct using
multi-class text classification on eight languages and eight models.
| 2,020 | Computation and Language |
Linguistic Resources for Bhojpuri, Magahi and Maithili: Statistics about
them, their Similarity Estimates, and Baselines for Three Applications | Corpus preparation for low-resource languages and for development of human
language technology to analyze or computationally process them is a laborious
task, primarily due to the unavailability of expert linguists who are native
speakers of these languages and also due to the time and resources required.
Bhojpuri, Magahi, and Maithili, languages of the Purvanchal region of India (in
the north-eastern parts), are low-resource languages belonging to the
Indo-Aryan (or Indic) family. They are closely related to Hindi, which is a
relatively high-resource language, which is why we compare with Hindi. We
collected corpora for these three languages from various sources and cleaned
them to the extent possible, without changing the data in them. The text
belongs to different domains and genres. We calculated some basic statistical
measures for these corpora at character, word, syllable, and morpheme levels.
These corpora were also annotated with parts-of-speech (POS) and chunk tags.
The basic statistical measures were both absolute and relative and were
exptected to indicate of linguistic properties such as morphological, lexical,
phonological, and syntactic complexities (or richness). The results were
compared with a standard Hindi corpus. For most of the measures, we tried to
the corpus size the same across the languages to avoid the effect of corpus
size, but in some cases it turned out that using the full corpus was better,
even if sizes were very different. Although the results are not very clear, we
try to draw some conclusions about the languages and the corpora. For POS
tagging and chunking, the BIS tagset was used to manually annotate the data.
The POS tagged data sizes are 16067, 14669 and 12310 sentences, respectively,
for Bhojpuri, Magahi and Maithili. The sizes for chunking are 9695 and 1954
sentences for Bhojpuri and Maithili, respectively.
| 2,021 | Computation and Language |
BURT: BERT-inspired Universal Representation from Twin Structure | Pre-trained contextualized language models such as BERT have shown great
effectiveness in a wide range of downstream Natural Language Processing (NLP)
tasks. However, the effective representations offered by the models target at
each token inside a sequence rather than each sequence and the fine-tuning step
involves the input of both sequences at one time, leading to unsatisfying
representations of various sequences with different granularities. Especially,
as sentence-level representations taken as the full training context in these
models, there comes inferior performance on lower-level linguistic units
(phrases and words). In this work, we present BURT (BERT inspired Universal
Representation from Twin Structure) that is capable of generating universal,
fixed-size representations for input sequences of any granularity, i.e., words,
phrases, and sentences, using a large scale of natural language inference and
paraphrase data with multiple training objectives. Our proposed BURT adopts the
Siamese network, learning sentence-level representations from natural language
inference dataset and word/phrase-level representations from paraphrasing
dataset, respectively. We evaluate BURT across different granularities of text
similarity tasks, including STS tasks, SemEval2013 Task 5(a) and some commonly
used word similarity tasks, where BURT substantially outperforms other
representation models on sentence-level datasets and achieves significant
improvements in word/phrase-level representation.
| 2,020 | Computation and Language |
Data Augmentation for Spoken Language Understanding via Pretrained
Language Models | The training of spoken language understanding (SLU) models often faces the
problem of data scarcity. In this paper, we put forward a data augmentation
method using pretrained language models to boost the variability and accuracy
of generated utterances. Furthermore, we investigate and propose solutions to
two previously overlooked semi-supervised learning scenarios of data scarcity
in SLU: i) Rich-in-Ontology: ontology information with numerous valid dialogue
acts is given; ii) Rich-in-Utterance: a large number of unlabelled utterances
are available. Empirical results show that our method can produce synthetic
training data that boosts the performance of language understanding models in
various scenarios.
| 2,021 | Computation and Language |
Zero-shot topic generation | We present an approach to generating topics using a model trained only for
document title generation, with zero examples of topics given during training.
We leverage features that capture the relevance of a candidate span in a
document for the generation of a title for that document. The output is a
weighted collection of the phrases that are most relevant for describing the
document and distinguishing it within a corpus, without requiring access to the
rest of the corpus. We conducted a double-blind trial in which human annotators
scored the quality of our machine-generated topics along with original
human-written topics associated with news articles from The Guardian and The
Huffington Post. The results show that our zero-shot model generates topic
labels for news documents that are on average equal to or higher quality than
those written by humans, as judged by humans.
| 2,020 | Computation and Language |
Measuring Information Propagation in Literary Social Networks | We present the task of modeling information propagation in literature, in
which we seek to identify pieces of information passing from character A to
character B to character C, only given a description of their activity in text.
We describe a new pipeline for measuring information propagation in this domain
and publish a new dataset for speaker attribution, enabling the evaluation of
an important component of this pipeline on a wider range of literary texts than
previously studied. Using this pipeline, we analyze the dynamics of information
propagation in over 5,000 works of fiction, finding that information flows
through characters that fill structural holes connecting different communities,
and that characters who are women are depicted as filling this role much more
frequently than characters who are men.
| 2,020 | Computation and Language |
Conditional Neural Generation using Sub-Aspect Functions for Extractive
News Summarization | Much progress has been made in text summarization, fueled by neural
architectures using large-scale training corpora. However, in the news domain,
neural models easily overfit by leveraging position-related features due to the
prevalence of the inverted pyramid writing style. In addition, there is an
unmet need to generate a variety of summaries for different users. In this
paper, we propose a neural framework that can flexibly control summary
generation by introducing a set of sub-aspect functions (i.e. importance,
diversity, position). These sub-aspect functions are regulated by a set of
control codes to decide which sub-aspect to focus on during summary generation.
We demonstrate that extracted summaries with minimal position bias is
comparable with those generated by standard models that take advantage of
position preference. We also show that news summaries generated with a focus on
diversity can be more preferred by human raters. These results suggest that a
more flexible neural summarization framework providing more control options
could be desirable in tailoring to different user preferences, which is useful
since it is often impractical to articulate such preferences for different
applications a priori.
| 2,020 | Computation and Language |
Knowledgeable Dialogue Reading Comprehension on Key Turns | Multi-choice machine reading comprehension (MRC) requires models to choose
the correct answer from candidate options given a passage and a question. Our
research focuses dialogue-based MRC, where the passages are multi-turn
dialogues. It suffers from two challenges, the answer selection decision is
made without support of latently helpful commonsense, and the multi-turn
context may hide considerable irrelevant information. This work thus makes the
first attempt to tackle those two challenges by extracting substantially
important turns and utilizing external knowledge to enhance the representation
of context. In this paper, the relevance of each turn to the question are
calculated to choose key turns. Besides, terms related to the context and the
question in a knowledge graph are extracted as external knowledge. The original
context, question and external knowledge are encoded with the pre-trained
language model, then the language representation and key turns are combined
together with a will-designed mechanism to predict the answer. Experimental
results on a DREAM dataset show that our proposed model achieves great
improvements on baselines.
| 2,020 | Computation and Language |
Benchmarking Robustness of Machine Reading Comprehension Models | Machine Reading Comprehension (MRC) is an important testbed for evaluating
models' natural language understanding (NLU) ability. There has been rapid
progress in this area, with new models achieving impressive performance on
various benchmarks. However, existing benchmarks only evaluate models on
in-domain test sets without considering their robustness under test-time
perturbations or adversarial attacks. To fill this important gap, we construct
AdvRACE (Adversarial RACE), a new model-agnostic benchmark for evaluating the
robustness of MRC models under four different types of adversarial attacks,
including our novel distractor extraction and generation attacks. We show that
state-of-the-art (SOTA) models are vulnerable to all of these attacks. We
conclude that there is substantial room for building more robust MRC models and
our benchmark can help motivate and measure progress in this area. We release
our data and code at https://github.com/NoviScl/AdvRACE .
| 2,021 | Computation and Language |
Filtering Noisy Dialogue Corpora by Connectivity and Content Relatedness | Large-scale dialogue datasets have recently become available for training
neural dialogue agents. However, these datasets have been reported to contain a
non-negligible number of unacceptable utterance pairs. In this paper, we
propose a method for scoring the quality of utterance pairs in terms of their
connectivity and relatedness. The proposed scoring method is designed based on
findings widely shared in the dialogue and linguistics research communities. We
demonstrate that it has a relatively good correlation with the human judgment
of dialogue quality. Furthermore, the method is applied to filter out
potentially unacceptable utterance pairs from a large-scale noisy dialogue
corpus to ensure its quality. We experimentally confirm that training data
filtered by the proposed method improves the quality of neural dialogue agents
in response generation.
| 2,020 | Computation and Language |
Multiscale Collaborative Deep Models for Neural Machine Translation | Recent evidence reveals that Neural Machine Translation (NMT) models with
deeper neural networks can be more effective but are difficult to train. In
this paper, we present a MultiScale Collaborative (MSC) framework to ease the
training of NMT models that are substantially deeper than those used
previously. We explicitly boost the gradient back-propagation from top to
bottom levels by introducing a block-scale collaboration mechanism into deep
NMT models. Then, instead of forcing the whole encoder stack directly learns a
desired representation, we let each encoder block learns a fine-grained
representation and enhance it by encoding spatial dependencies using a
context-scale collaboration. We provide empirical evidence showing that the MSC
nets are easy to optimize and can obtain improvements of translation quality
from considerably increased depth. On IWSLT translation tasks with three
translation directions, our extremely deep models (with 72-layer encoders)
surpass strong baselines by +2.2~+3.1 BLEU points. In addition, our deep MSC
achieves a BLEU score of 30.56 on WMT14 English-German task that significantly
outperforms state-of-the-art deep NMT models.
| 2,020 | Computation and Language |
Morphological Disambiguation of South S\'ami with FSTs and Neural
Networks | We present a method for conducting morphological disambiguation for South
S\'ami, which is an endangered language. Our method uses an FST-based
morphological analyzer to produce an ambiguous set of morphological readings
for each word in a sentence. These readings are disambiguated with a Bi-RNN
model trained on the related North S\'ami UD Treebank and some synthetically
generated South S\'ami data. The disambiguation is done on the level of
morphological tags ignoring word forms and lemmas; this makes it possible to
use North S\'ami training data for South S\'ami without the need for a
bilingual dictionary or aligned word embeddings. Our approach requires only
minimal resources for South S\'ami, which makes it usable and applicable in the
contexts of any other endangered language as well.
| 2,020 | Computation and Language |
Automatically Identifying Gender Issues in Machine Translation using
Perturbations | The successful application of neural methods to machine translation has
realized huge quality advances for the community. With these improvements, many
have noted outstanding challenges, including the modeling and treatment of
gendered language. While previous studies have identified issues using
synthetic examples, we develop a novel technique to mine examples from real
world data to explore challenges for deployed systems. We use our method to
compile an evaluation benchmark spanning examples for four languages from three
language families, which we publicly release to facilitate research. The
examples in our benchmark expose where model representations are gendered, and
the unintended consequences these gendered representations can have in
downstream application.
| 2,020 | Computation and Language |
Enhancing Answer Boundary Detection for Multilingual Machine Reading
Comprehension | Multilingual pre-trained models could leverage the training data from a rich
source language (such as English) to improve performance on low resource
languages. However, the transfer quality for multilingual Machine Reading
Comprehension (MRC) is significantly worse than sentence classification tasks
mainly due to the requirement of MRC to detect the word level answer boundary.
In this paper, we propose two auxiliary tasks in the fine-tuning stage to
create additional phrase boundary supervision: (1) A mixed MRC task, which
translates the question or passage to other languages and builds cross-lingual
question-passage pairs; (2) A language-agnostic knowledge masking task by
leveraging knowledge phrases mined from web. Besides, extensive experiments on
two cross-lingual MRC datasets show the effectiveness of our proposed approach.
| 2,020 | Computation and Language |
Pre-training Is (Almost) All You Need: An Application to Commonsense
Reasoning | Fine-tuning of pre-trained transformer models has become the standard
approach for solving common NLP tasks. Most of the existing approaches rely on
a randomly initialized classifier on top of such networks. We argue that this
fine-tuning procedure is sub-optimal as the pre-trained model has no prior on
the specific classifier labels, while it might have already learned an
intrinsic textual representation of the task. In this paper, we introduce a new
scoring method that casts a plausibility ranking task in a full-text format and
leverages the masked language modeling head tuned during the pre-training
phase. We study commonsense reasoning tasks where the model must rank a set of
hypotheses given a premise, focusing on the COPA, Swag, HellaSwag and
CommonsenseQA datasets. By exploiting our scoring method without fine-tuning,
we are able to produce strong baselines (e.g. 80% test accuracy on COPA) that
are comparable to supervised approaches. Moreover, when fine-tuning directly on
the proposed scoring function, we show that our method provides a much more
stable training phase across random restarts (e.g $\times 10$ standard
deviation reduction on COPA test accuracy) and requires less annotated data
than the standard classifier approach to reach equivalent performances.
| 2,020 | Computation and Language |
Modeling Long Context for Task-Oriented Dialogue State Generation | Based on the recently proposed transferable dialogue state generator (TRADE)
that predicts dialogue states from utterance-concatenated dialogue context, we
propose a multi-task learning model with a simple yet effective utterance
tagging technique and a bidirectional language model as an auxiliary task for
task-oriented dialogue state generation. By enabling the model to learn a
better representation of the long dialogue context, our approaches attempt to
solve the problem that the performance of the baseline significantly drops when
the input dialogue context sequence is long. In our experiments, our proposed
model achieves a 7.03% relative improvement over the baseline, establishing a
new state-of-the-art joint goal accuracy of 52.04% on the MultiWOZ 2.0 dataset.
| 2,020 | Computation and Language |
Demographics Should Not Be the Reason of Toxicity: Mitigating
Discrimination in Text Classifications with Instance Weighting | With the recent proliferation of the use of text classifications, researchers
have found that there are certain unintended biases in text classification
datasets. For example, texts containing some demographic identity-terms (e.g.,
"gay", "black") are more likely to be abusive in existing abusive language
detection datasets. As a result, models trained with these datasets may
consider sentences like "She makes me happy to be gay" as abusive simply
because of the word "gay." In this paper, we formalize the unintended biases in
text classification datasets as a kind of selection bias from the
non-discrimination distribution to the discrimination distribution. Based on
this formalization, we further propose a model-agnostic debiasing training
framework by recovering the non-discrimination distribution using instance
weighting, which does not require any extra resources or annotations apart from
a pre-defined set of demographic identity-terms. Experiments demonstrate that
our method can effectively alleviate the impacts of the unintended biases
without significantly hurting models' generalization ability.
| 2,020 | Computation and Language |
Do Neural Language Models Show Preferences for Syntactic Formalisms? | Recent work on the interpretability of deep neural language models has
concluded that many properties of natural language syntax are encoded in their
representational spaces. However, such studies often suffer from limited scope
by focusing on a single language and a single linguistic formalism. In this
study, we aim to investigate the extent to which the semblance of syntactic
structure captured by language models adheres to a surface-syntactic or deep
syntactic style of analysis, and whether the patterns are consistent across
different languages. We apply a probe for extracting directed dependency trees
to BERT and ELMo models trained on 13 different languages, probing for two
different syntactic annotation styles: Universal Dependencies (UD),
prioritizing deep syntactic relations, and Surface-Syntactic Universal
Dependencies (SUD), focusing on surface structure. We find that both models
exhibit a preference for UD over SUD - with interesting variations across
languages and layers - and that the strength of this preference is correlated
with differences in tree shape.
| 2,020 | Computation and Language |
Adversarial Subword Regularization for Robust Neural Machine Translation | Exposing diverse subword segmentations to neural machine translation (NMT)
models often improves the robustness of machine translation as NMT models can
experience various subword candidates. However, the diversification of subword
segmentations mostly relies on the pre-trained subword language models from
which erroneous segmentations of unseen words are less likely to be sampled. In
this paper, we present adversarial subword regularization (ADVSR) to study
whether gradient signals during training can be a substitute criterion for
exposing diverse subword segmentations. We experimentally show that our
model-based adversarial samples effectively encourage NMT models to be less
sensitive to segmentation errors and improve the performance of NMT models in
low-resource and out-domain datasets.
| 2,020 | Computation and Language |
Analysing Lexical Semantic Change with Contextualised Word
Representations | This paper presents the first unsupervised approach to lexical semantic
change that makes use of contextualised word representations. We propose a
novel method that exploits the BERT neural language model to obtain
representations of word usages, clusters these representations into usage
types, and measures change along time with three proposed metrics. We create a
new evaluation dataset and show that the model representations and the detected
semantic shifts are positively correlated with human judgements. Our extensive
qualitative analysis demonstrates that our method captures a variety of
synchronic and diachronic linguistic phenomena. We expect our work to inspire
further research in this direction.
| 2,020 | Computation and Language |
Combining Word Embeddings and N-grams for Unsupervised Document
Summarization | Graph-based extractive document summarization relies on the quality of the
sentence similarity graph. Bag-of-words or tf-idf based sentence similarity
uses exact word matching, but fails to measure the semantic similarity between
individual words or to consider the semantic structure of sentences. In order
to improve the similarity measure between sentences, we employ off-the-shelf
deep embedding features and tf-idf features, and introduce a new text
similarity metric. An improved sentence similarity graph is built and used in a
submodular objective function for extractive summarization, which consists of a
weighted coverage term and a diversity term. A Transformer based compression
model is developed for sentence compression to aid in document summarization.
Our summarization approach is extractive and unsupervised. Experiments
demonstrate that our approach can outperform the tf-idf based approach and
achieve state-of-the-art performance on the DUC04 dataset, and comparable
performance to the fully supervised learning methods on the CNN/DM and NYT
datasets.
| 2,020 | Computation and Language |
Learning Non-Monotonic Automatic Post-Editing of Translations from Human
Orderings | Recent research in neural machine translation has explored flexible
generation orders, as an alternative to left-to-right generation. However,
training non-monotonic models brings a new complication: how to search for a
good ordering when there is a combinatorial explosion of orderings arriving at
the same final result? Also, how do these automatic orderings compare with the
actual behaviour of human translators? Current models rely on manually built
biases or are left to explore all possibilities on their own. In this paper, we
analyze the orderings produced by human post-editors and use them to train an
automatic post-editing system. We compare the resulting system with those
trained with left-to-right and random post-editing orderings. We observe that
humans tend to follow a nearly left-to-right order, but with interesting
deviations, such as preferring to start by correcting punctuation or verbs.
| 2,021 | Computation and Language |
How fine can fine-tuning be? Learning efficient language models | State-of-the-art performance on language understanding tasks is now achieved
with increasingly large networks; the current record holder has billions of
parameters. Given a language model pre-trained on massive unlabeled text
corpora, only very light supervised fine-tuning is needed to learn a task: the
number of fine-tuning steps is typically five orders of magnitude lower than
the total parameter count. Does this mean that fine-tuning only introduces
small differences from the pre-trained model in the parameter space? If so, can
one avoid storing and computing an entire model for each task? In this work, we
address these questions by using Bidirectional Encoder Representations from
Transformers (BERT) as an example. As expected, we find that the fine-tuned
models are close in parameter space to the pre-trained one, with the closeness
varying from layer to layer. We show that it suffices to fine-tune only the
most critical layers. Further, we find that there are surprisingly many good
solutions in the set of sparsified versions of the pre-trained model. As a
result, fine-tuning of huge language models can be achieved by simply setting a
certain number of entries in certain layers of the pre-trained parameters to
zero, saving both task-specific parameter storage and computational cost.
| 2,020 | Computation and Language |
A Workflow Manager for Complex NLP and Content Curation Pipelines | We present a workflow manager for the flexible creation and customisation of
NLP processing pipelines. The workflow manager addresses challenges in
interoperability across various different NLP tasks and hardware-based resource
usage. Based on the four key principles of generality, flexibility, scalability
and efficiency, we present the first version of the workflow manager by
providing details on its custom definition language, explaining the
communication components and the general system architecture and setup. We
currently implement the system, which is grounded and motivated by real-world
industry use cases in several innovation and transfer projects.
| 2,020 | Computation and Language |
Using Punkt for Sentence Segmentation in non-Latin Scripts: Experiments
on Kurdish (Sorani) Texts | Segmentation is a fundamental step for most Natural Language Processing
tasks. The Kurdish language is a multi-dialect, under-resourced language which
is written in different scripts. The lack of various segmented corpora is one
of the major bottlenecks in Kurdish language processing. We used Punkt, an
unsupervised machine learning method, to segment a Kurdish corpus of Sorani
dialect, written in Persian-Arabic script. According to the literature, studies
on using Punkt on non-Latin data are scanty. In our experiment, we achieved an
F1 score of 91.10% and had an Error Rate of 16.32%. The high Error Rate is
mainly due to the situation of abbreviations in Kurdish and partly because of
ordinal numerals. The data is publicly available at
https://github.com/KurdishBLARK/ KTC-Segmented for non-commercial use under the
CC BY-NC-SA 4.0 licence.
| 2,020 | Computation and Language |
BERT Fine-tuning For Arabic Text Summarization | Fine-tuning a pretrained BERT model is the state of the art method for
extractive/abstractive text summarization, in this paper we showcase how this
fine-tuning method can be applied to the Arabic language to both construct the
first documented model for abstractive Arabic text summarization and show its
performance in Arabic extractive summarization. Our model works with
multilingual BERT (as Arabic language does not have a pretrained BERT of its
own). We show its performance in English corpus first before applying it to
Arabic corpora in both extractive and abstractive tasks.
| 2,020 | Computation and Language |
Entity Candidate Network for Whole-Aware Named Entity Recognition | Named Entity Recognition (NER) is a crucial upstream task in Natural Language
Processing (NLP). Traditional tag scheme approaches offer a single recognition
that does not meet the needs of many downstream tasks such as coreference
resolution. Meanwhile, Tag scheme approaches ignore the continuity of entities.
Inspired by one-stage object detection models in computer vision (CV), this
paper proposes a new no-tag scheme, the Whole-Aware Detection, which makes NER
an object detection task. Meanwhile, this paper presents a novel model, Entity
Candidate Network (ECNet), and a specific convolution network, Adaptive Context
Convolution Network (ACCN), to fuse multi-scale contexts and encode entity
information at each position. ECNet identifies the full span of a named entity
and its type at each position based on Entity Loss. Furthermore, ECNet is
regulable between the highest precision and the highest recall, while the tag
scheme approaches are not. Experimental results on the CoNLL 2003 English
dataset and the WNUT 2017 dataset show that ECNet outperforms other previous
state-of-the-art methods.
| 2,020 | Computation and Language |
MICK: A Meta-Learning Framework for Few-shot Relation Classification
with Small Training Data | Few-shot relation classification seeks to classify incoming query instances
after meeting only few support instances. This ability is gained by training
with large amount of in-domain annotated data. In this paper, we tackle an even
harder problem by further limiting the amount of data available at training
time. We propose a few-shot learning framework for relation classification,
which is particularly powerful when the training data is very small. In this
framework, models not only strive to classify query instances, but also seek
underlying knowledge about the support instances to obtain better instance
representations. The framework also includes a method for aggregating
cross-domain knowledge into models by open-source task enrichment.
Additionally, we construct a brand new dataset: the TinyRel-CM dataset, a
few-shot relation classification dataset in health domain with purposely small
training data and challenging relation classes. Experimental results
demonstrate that our framework brings performance gains for most underlying
classification models, outperforms the state-of-the-art results given small
training data, and achieves competitive results with sufficiently large
training data.
| 2,020 | Computation and Language |
Classification of Cuisines from Sequentially Structured Recipes | Cultures across the world are distinguished by the idiosyncratic patterns in
their cuisines. These cuisines are characterized in terms of their
substructures such as ingredients, cooking processes and utensils. A complex
fusion of these substructures intrinsic to a region defines the identity of a
cuisine. Accurate classification of cuisines based on their culinary features
is an outstanding problem and has hitherto been attempted to solve by
accounting for ingredients of a recipe as features. Previous studies have
attempted cuisine classification by using unstructured recipes without
accounting for details of cooking techniques. In reality, the cooking
processes/techniques and their order are highly significant for the recipe's
structure and hence for its classification. In this article, we have
implemented a range of classification techniques by accounting for this
information on the RecipeDB dataset containing sequential data on recipes. The
state-of-the-art RoBERTa model presented the highest accuracy of 73.30% among a
range of classification models from Logistic Regression and Naive Bayes to
LSTMs and Transformers.
| 2,020 | Computation and Language |
SpellGCN: Incorporating Phonological and Visual Similarities into
Language Models for Chinese Spelling Check | Chinese Spelling Check (CSC) is a task to detect and correct spelling errors
in Chinese natural language. Existing methods have made attempts to incorporate
the similarity knowledge between Chinese characters. However, they take the
similarity knowledge as either an external input resource or just heuristic
rules. This paper proposes to incorporate phonological and visual similarity
knowledge into language models for CSC via a specialized graph convolutional
network (SpellGCN). The model builds a graph over the characters, and SpellGCN
is learned to map this graph into a set of inter-dependent character
classifiers. These classifiers are applied to the representations extracted by
another network, such as BERT, enabling the whole network to be end-to-end
trainable. Experiments (The dataset and all code for this paper are available
at https://github.com/ACL2020SpellGCN/SpellGCN) are conducted on three
human-annotated datasets. Our method achieves superior performance against
previous models by a large margin.
| 2,020 | Computation and Language |
Reevaluating Adversarial Examples in Natural Language | State-of-the-art attacks on NLP models lack a shared definition of a what
constitutes a successful attack. We distill ideas from past work into a unified
framework: a successful natural language adversarial example is a perturbation
that fools the model and follows some linguistic constraints. We then analyze
the outputs of two state-of-the-art synonym substitution attacks. We find that
their perturbations often do not preserve semantics, and 38% introduce
grammatical errors. Human surveys reveal that to successfully preserve
semantics, we need to significantly increase the minimum cosine similarities
between the embeddings of swapped words and between the sentence encodings of
original and perturbed sentences.With constraints adjusted to better preserve
semantics and grammaticality, the attack success rate drops by over 70
percentage points.
| 2,020 | Computation and Language |
Development of a General Purpose Sentiment Lexicon for Igbo Language | There are publicly available general purpose sentiment lexicons in some high
resource languages but very few exist in the low resource languages. This makes
it difficult to directly perform sentiment analysis tasks in such languages.
The objective of this work is to create a general purpose sentiment lexicon for
the Igbo language that can determine the sentiment of documents written in the
Igbo language without having to translate it to the English language. The
material used was an automatically translated lexicon by Liu and the manual
addition of Igbo native words. The result of this work is a general purpose
lexicon called IgboSentilex. The performance was tested on the BBC Igbo news
channel. It returned an average polarity agreement of 95.75 percent with other
general purpose sentiment lexicons.
| 2,020 | Computation and Language |
Multimodal Routing: Improving Local and Global Interpretability of
Multimodal Language Analysis | The human language can be expressed through multiple sources of information
known as modalities, including tones of voice, facial gestures, and spoken
language. Recent multimodal learning with strong performances on human-centric
tasks such as sentiment analysis and emotion recognition are often black-box,
with very limited interpretability. In this paper we propose Multimodal
Routing, which dynamically adjusts weights between input modalities and output
representations differently for each input sample. Multimodal routing can
identify relative importance of both individual modalities and cross-modality
features. Moreover, the weight assignment by routing allows us to interpret
modality-prediction relationships not only globally (i.e. general trends over
the whole dataset), but also locally for each single input sample, meanwhile
keeping competitive performance compared to state-of-the-art methods.
| 2,020 | Computation and Language |
Syntax-aware Data Augmentation for Neural Machine Translation | Data augmentation is an effective performance enhancement in neural machine
translation (NMT) by generating additional bilingual data. In this paper, we
propose a novel data augmentation enhancement strategy for neural machine
translation. Different from existing data augmentation methods which simply
choose words with the same probability across different sentences for
modification, we set sentence-specific probability for word selection by
considering their roles in sentence. We use dependency parse tree of input
sentence as an effective clue to determine selecting probability for every
words in each sentence. Our proposed method is evaluated on WMT14
English-to-German dataset and IWSLT14 German-to-English dataset. The result of
extensive experiments show our proposed syntax-aware data augmentation method
may effectively boost existing sentence-independent methods for significant
translation performance improvement.
| 2,020 | Computation and Language |
Leveraging Declarative Knowledge in Text and First-Order Logic for
Fine-Grained Propaganda Detection | We study the detection of propagandistic text fragments in news articles.
Instead of merely learning from input-output datapoints in training data, we
introduce an approach to inject declarative knowledge of fine-grained
propaganda techniques. Specifically, we leverage the declarative knowledge
expressed in both first-order logic and natural language. The former refers to
the logical consistency between coarse- and fine-grained predictions, which is
used to regularize the training process with propositional Boolean expressions.
The latter refers to the literal definition of each propaganda technique, which
is utilized to get class representations for regularizing the model parameters.
We conduct experiments on Propaganda Techniques Corpus, a large manually
annotated dataset for fine-grained propaganda detection. Experiments show that
our method achieves superior performance, demonstrating that leveraging
declarative knowledge can help the model to make more accurate predictions.
| 2,020 | Computation and Language |
Exploring Fine-tuning Techniques for Pre-trained Cross-lingual Models
via Continual Learning | Recently, fine-tuning pre-trained language models (e.g., multilingual BERT)
to downstream cross-lingual tasks has shown promising results. However, the
fine-tuning process inevitably changes the parameters of the pre-trained model
and weakens its cross-lingual ability, which leads to sub-optimal performance.
To alleviate this problem, we leverage continual learning to preserve the
original cross-lingual ability of the pre-trained model when we fine-tune it to
downstream tasks. The experimental result shows that our fine-tuning methods
can better preserve the cross-lingual ability of the pre-trained model in a
sentence retrieval task. Our methods also achieve better performance than other
fine-tuning baselines on the zero-shot cross-lingual part-of-speech tagging and
named entity recognition tasks.
| 2,020 | Computation and Language |
Exploiting Structured Knowledge in Text via Graph-Guided Representation
Learning | In this work, we aim at equipping pre-trained language models with structured
knowledge. We present two self-supervised tasks learning over raw text with the
guidance from knowledge graphs. Building upon entity-level masked language
models, our first contribution is an entity masking scheme that exploits
relational knowledge underlying the text. This is fulfilled by using a linked
knowledge graph to select informative entities and then masking their mentions.
In addition we use knowledge graphs to obtain distractors for the masked
entities, and propose a novel distractor-suppressed ranking objective which is
optimized jointly with masked language model. In contrast to existing
paradigms, our approach uses knowledge graphs implicitly, only during
pre-training, to inject language models with structured knowledge via learning
from raw text. It is more efficient than retrieval-based methods that perform
entity linking and integration during finetuning and inference, and generalizes
more effectively than the methods that directly learn from concatenated graph
triples. Experiments show that our proposed model achieves improved performance
on five benchmark datasets, including question answering and knowledge base
completion tasks.
| 2,020 | Computation and Language |
Meta-Transfer Learning for Code-Switched Speech Recognition | An increasing number of people in the world today speak a mixed-language as a
result of being multilingual. However, building a speech recognition system for
code-switching remains difficult due to the availability of limited resources
and the expense and significant effort required to collect mixed-language data.
We therefore propose a new learning method, meta-transfer learning, to transfer
learn on a code-switched speech recognition system in a low-resource setting by
judiciously extracting information from high-resource monolingual datasets. Our
model learns to recognize individual languages, and transfer them so as to
better recognize mixed-language speech by conditioning the optimization on the
code-switching data. Based on experimental results, our model outperforms
existing baselines on speech recognition and language modeling tasks, and is
faster to converge.
| 2,020 | Computation and Language |
Normalizing Compositional Structures Across Graphbanks | The emergence of a variety of graph-based meaning representations (MRs) has
sparked an important conversation about how to adequately represent semantic
structure. These MRs exhibit structural differences that reflect different
theoretical and design considerations, presenting challenges to uniform
linguistic analysis and cross-framework semantic parsing. Here, we ask the
question of which design differences between MRs are meaningful and
semantically-rooted, and which are superficial. We present a methodology for
normalizing discrepancies between MRs at the compositional level (Lindemann et
al., 2019), finding that we can normalize the majority of divergent phenomena
using linguistically-grounded rules. Our work significantly increases the match
in compositional structure between MRs and improves multi-task learning (MTL)
in a low-resource setting, demonstrating the usefulness of careful MR design
analysis and comparison.
| 2,020 | Computation and Language |
Towards Transparent and Explainable Attention Models | Recent studies on interpretability of attention distributions have led to
notions of faithful and plausible explanations for a model's predictions.
Attention distributions can be considered a faithful explanation if a higher
attention weight implies a greater impact on the model's prediction. They can
be considered a plausible explanation if they provide a human-understandable
justification for the model's predictions. In this work, we first explain why
current attention mechanisms in LSTM based encoders can neither provide a
faithful nor a plausible explanation of the model's predictions. We observe
that in LSTM based encoders the hidden representations at different time-steps
are very similar to each other (high conicity) and attention weights in these
situations do not carry much meaning because even a random permutation of the
attention weights does not affect the model's predictions. Based on experiments
on a wide variety of tasks and datasets, we observe attention distributions
often attribute the model's predictions to unimportant words such as
punctuation and fail to offer a plausible explanation for the predictions. To
make attention mechanisms more faithful and plausible, we propose a modified
LSTM cell with a diversity-driven training objective that ensures that the
hidden representations learned at different time steps are diverse. We show
that the resulting attention distributions offer more transparency as they (i)
provide a more precise importance ranking of the hidden states (ii) are better
indicative of words important for the model's predictions (iii) correlate
better with gradient-based attribution methods. Human evaluations indicate that
the attention distributions learned by our model offer a plausible explanation
of the model's predictions. Our code has been made publicly available at
https://github.com/akashkm99/Interpretable-Attention
| 2,020 | Computation and Language |
GePpeTto Carves Italian into a Language Model | In the last few years, pre-trained neural architectures have provided
impressive improvements across several NLP tasks. Still, generative language
models are available mainly for English. We develop GePpeTto, the first
generative language model for Italian, built using the GPT-2 architecture. We
provide a thorough analysis of GePpeTto's quality by means of both an automatic
and a human-based evaluation. The automatic assessment consists in (i)
calculating perplexity across different genres and (ii) a profiling analysis
over GePpeTto's writing characteristics. We find that GePpeTto's production is
a sort of bonsai version of human production, with shorter but yet complex
sentences. Human evaluation is performed over a sentence completion task, where
GePpeTto's output is judged as natural more often than not, and much closer to
the original human texts than to a simpler language model which we take as
baseline.
| 2,020 | Computation and Language |
Politeness Transfer: A Tag and Generate Approach | This paper introduces a new task of politeness transfer which involves
converting non-polite sentences to polite sentences while preserving the
meaning. We also provide a dataset of more than 1.39 instances automatically
labeled for politeness to encourage benchmark evaluations on this new task. We
design a tag and generate pipeline that identifies stylistic attributes and
subsequently generates a sentence in the target style while preserving most of
the source content. For politeness as well as five other transfer tasks, our
model outperforms the state-of-the-art methods on automatic metrics for content
preservation, with a comparable or better performance on style transfer
accuracy. Additionally, our model surpasses existing methods on human
evaluations for grammaticality, meaning preservation and transfer accuracy
across all the six style transfer tasks. The data and code is located at
https://github.com/tag-and-generate.
| 2,020 | Computation and Language |
Exploring the Suitability of Semantic Spaces as Word Association Models
for the Extraction of Semantic Relationships | Given the recent advances and progress in Natural Language Processing (NLP),
extraction of semantic relationships has been at the top of the research agenda
in the last few years. This work has been mainly motivated by the fact that
building knowledge graphs (KG) and bases (KB), as a key ingredient of
intelligent applications, is a never-ending challenge, since new knowledge
needs to be harvested while old knowledge needs to be revised. Currently,
approaches towards relation extraction from text are dominated by neural models
practicing some sort of distant (weak) supervision in machine learning from
large corpora, with or without consulting external knowledge sources. In this
paper, we empirically study and explore the potential of a novel idea of using
classical semantic spaces and models, e.g., Word Embedding, generated for
extracting word association, in conjunction with relation extraction
approaches. The goal is to use these word association models to reinforce
current relation extraction approaches. We believe that this is a first attempt
of this kind and the results of the study should shed some light on the extent
to which these word association models can be used as well as the most
promising types of relationships to be considered for extraction.
| 2,020 | Computation and Language |
Towards Reasonably-Sized Character-Level Transformer NMT by Finetuning
Subword Systems | Applying the Transformer architecture on the character level usually requires
very deep architectures that are difficult and slow to train. These problems
can be partially overcome by incorporating a segmentation into tokens in the
model. We show that by initially training a subword model and then finetuning
it on characters, we can obtain a neural machine translation model that works
at the character level without requiring token segmentation. We use only the
vanilla 6-layer Transformer Base architecture. Our character-level models
better capture morphological phenomena and show more robustness to noise at the
expense of somewhat worse overall translation quality. Our study is a
significant step towards high-performance and easy to train character-based
models that are not extremely large.
| 2,020 | Computation and Language |
SubjQA: A Dataset for Subjectivity and Review Comprehension | Subjectivity is the expression of internal opinions or beliefs which cannot
be objectively observed or verified, and has been shown to be important for
sentiment analysis and word-sense disambiguation. Furthermore, subjectivity is
an important aspect of user-generated data. In spite of this, subjectivity has
not been investigated in contexts where such data is widespread, such as in
question answering (QA). We therefore investigate the relationship between
subjectivity and QA, while developing a new dataset. We compare and contrast
with analyses from previous work, and verify that findings regarding
subjectivity still hold when using recently developed NLP architectures. We
find that subjectivity is also an important feature in the case of QA, albeit
with more intricate interactions between subjectivity and QA performance. For
instance, a subjective question may or may not be associated with a subjective
answer. We release an English QA dataset (SubjQA) based on customer reviews,
containing subjectivity annotations for questions and answer spans across 6
distinct domains.
| 2,020 | Computation and Language |
General Purpose Text Embeddings from Pre-trained Language Models for
Scalable Inference | The state of the art on many NLP tasks is currently achieved by large
pre-trained language models, which require a considerable amount of
computation. We explore a setting where many different predictions are made on
a single piece of text. In that case, some of the computational cost during
inference can be amortized over the different tasks using a shared text
encoder. We compare approaches for training such an encoder and show that
encoders pre-trained over multiple tasks generalize well to unseen tasks. We
also compare ways of extracting fixed- and limited-size representations from
this encoder, including different ways of pooling features extracted from
multiple layers or positions. Our best approach compares favorably to knowledge
distillation, achieving higher accuracy and lower computational cost once the
system is handling around 7 tasks. Further, we show that through binary
quantization, we can reduce the size of the extracted representations by a
factor of 16 making it feasible to store them for later use. The resulting
method offers a compelling solution for using large-scale pre-trained models at
a fraction of the computational cost when multiple tasks are performed on the
same text.
| 2,020 | Computation and Language |
Detecting Perceived Emotions in Hurricane Disasters | Natural disasters (e.g., hurricanes) affect millions of people each year,
causing widespread destruction in their wake. People have recently taken to
social media websites (e.g., Twitter) to share their sentiments and feelings
with the larger community. Consequently, these platforms have become
instrumental in understanding and perceiving emotions at scale. In this paper,
we introduce HurricaneEmo, an emotion dataset of 15,000 English tweets spanning
three hurricanes: Harvey, Irma, and Maria. We present a comprehensive study of
fine-grained emotions and propose classification tasks to discriminate between
coarse-grained emotion groups. Our best BERT model, even after task-guided
pre-training which leverages unlabeled Twitter data, achieves only 68% accuracy
(averaged across all groups). HurricaneEmo serves not only as a challenging
benchmark for models but also as a valuable resource for analyzing emotions in
disaster-centric domains.
| 2,020 | Computation and Language |
Evaluating Dialogue Generation Systems via Response Selection | Existing automatic evaluation metrics for open-domain dialogue response
generation systems correlate poorly with human evaluation. We focus on
evaluating response generation systems via response selection. To evaluate
systems properly via response selection, we propose the method to construct
response selection test sets with well-chosen false candidates. Specifically,
we propose to construct test sets filtering out some types of false candidates:
(i) those unrelated to the ground-truth response and (ii) those acceptable as
appropriate responses. Through experiments, we demonstrate that evaluating
systems via response selection with the test sets developed by our method
correlates more strongly with human evaluation, compared with widely used
automatic evaluation metrics such as BLEU.
| 2,020 | Computation and Language |
TUNIZI: a Tunisian Arabizi sentiment analysis Dataset | On social media, Arabic people tend to express themselves in their own local
dialects. More particularly, Tunisians use the informal way called "Tunisian
Arabizi". Analytical studies seek to explore and recognize online opinions
aiming to exploit them for planning and prediction purposes such as measuring
the customer satisfaction and establishing sales and marketing strategies.
However, analytical studies based on Deep Learning are data hungry. On the
other hand, African languages and dialects are considered low resource
languages. For instance, to the best of our knowledge, no annotated Tunisian
Arabizi dataset exists. In this paper, we introduce TUNIZI a sentiment analysis
Tunisian Arabizi Dataset, collected from social networks, preprocessed for
analytical studies and annotated manually by Tunisian native speakers.
| 2,020 | Computation and Language |
UniConv: A Unified Conversational Neural Architecture for Multi-domain
Task-oriented Dialogues | Building an end-to-end conversational agent for multi-domain task-oriented
dialogues has been an open challenge for two main reasons. First, tracking
dialogue states of multiple domains is non-trivial as the dialogue agent must
obtain complete states from all relevant domains, some of which might have
shared slots among domains as well as unique slots specifically for one domain
only. Second, the dialogue agent must also process various types of information
across domains, including dialogue context, dialogue states, and database, to
generate natural responses to users. Unlike the existing approaches that are
often designed to train each module separately, we propose "UniConv" -- a novel
unified neural architecture for end-to-end conversational systems in
multi-domain task-oriented dialogues, which is designed to jointly train (i) a
Bi-level State Tracker which tracks dialogue states by learning signals at both
slot and domain level independently, and (ii) a Joint Dialogue Act and Response
Generator which incorporates information from various input components and
models dialogue acts and target responses simultaneously. We conduct
comprehensive experiments in dialogue state tracking, context-to-text, and
end-to-end settings on the MultiWOZ2.1 benchmark, achieving superior
performance over competitive baselines.
| 2,020 | Computation and Language |
A Cross-Genre Ensemble Approach to Robust Reddit Part of Speech Tagging | Part of speech tagging is a fundamental NLP task often regarded as solved for
high-resource languages such as English. Current state-of-the-art models have
achieved high accuracy, especially on the news domain. However, when these
models are applied to other corpora with different genres, and especially
user-generated data from the Web, we see substantial drops in performance. In
this work, we study how a state-of-the-art tagging model trained on different
genres performs on Web content from unfiltered Reddit forum discussions. More
specifically, we use data from multiple sources: OntoNotes, a large benchmark
corpus with 'well-edited' text, the English Web Treebank with 5 Web genres, and
GUM, with 7 further genres other than Reddit. We report the results when
training on different splits of the data, tested on Reddit. Our results show
that even small amounts of in-domain data can outperform the contribution of
data an order of magnitude larger coming from other Web domains. To make
progress on out-of-domain tagging, we also evaluate an ensemble approach using
multiple single-genre taggers as input features to a meta-classifier. We
present state of the art performance on tagging Reddit data, as well as error
analysis of the results of these models, and offer a typology of the most
common error types among them, broken down by training corpus.
| 2,020 | Computation and Language |
Don't Neglect the Obvious: On the Role of Unambiguous Words in Word
Sense Disambiguation | State-of-the-art methods for Word Sense Disambiguation (WSD) combine two
different features: the power of pre-trained language models and a propagation
method to extend the coverage of such models. This propagation is needed as
current sense-annotated corpora lack coverage of many instances in the
underlying sense inventory (usually WordNet). At the same time, unambiguous
words make for a large portion of all words in WordNet, while being poorly
covered in existing sense-annotated corpora. In this paper, we propose a simple
method to provide annotations for most unambiguous words in a large corpus. We
introduce the UWA (Unambiguous Word Annotations) dataset and show how a
state-of-the-art propagation-based model can use it to extend the coverage and
quality of its word sense embeddings by a significant margin, improving on its
original results on WSD.
| 2,020 | Computation and Language |
UDapter: Language Adaptation for Truly Universal Dependency Parsing | Recent advances in multilingual dependency parsing have brought the idea of a
truly universal parser closer to reality. However, cross-language interference
and restrained model capacity remain major obstacles. To address this, we
propose a novel multilingual task adaptation approach based on contextual
parameter generation and adapter modules. This approach enables to learn
adapters via language embeddings while sharing model parameters across
languages. It also allows for an easy but effective integration of existing
linguistic typology features into the parsing network. The resulting parser,
UDapter, outperforms strong monolingual and multilingual baselines on the
majority of both high-resource and low-resource (zero-shot) languages, showing
the success of the proposed adaptation approach. Our in-depth analyses show
that soft parameter sharing via typological features is key to this success.
| 2,020 | Computation and Language |
Beyond Instructional Videos: Probing for More Diverse Visual-Textual
Grounding on YouTube | Pretraining from unlabelled web videos has quickly become the de-facto means
of achieving high performance on many video understanding tasks. Features are
learned via prediction of grounded relationships between visual content and
automatic speech recognition (ASR) tokens. However, prior pretraining work has
been limited to only instructional videos; a priori, we expect this domain to
be relatively "easy:" speakers in instructional videos will often reference the
literal objects/actions being depicted. We ask: can similar models be trained
on more diverse video corpora? And, if so, what types of videos are "grounded"
and what types are not? We fit a representative pretraining model to the
diverse YouTube8M dataset, and study its success and failure cases. We find
that visual-textual grounding is indeed possible across previously unexplored
video categories, and that pretraining on a more diverse set results in
representations that generalize to both non-instructional and instructional
domains.
| 2,020 | Computation and Language |
End-to-End Slot Alignment and Recognition for Cross-Lingual NLU | Natural language understanding (NLU) in the context of goal-oriented dialog
systems typically includes intent classification and slot labeling tasks.
Existing methods to expand an NLU system to new languages use machine
translation with slot label projection from source to the translated
utterances, and thus are sensitive to projection errors. In this work, we
propose a novel end-to-end model that learns to align and predict target slot
labels jointly for cross-lingual transfer. We introduce MultiATIS++, a new
multilingual NLU corpus that extends the Multilingual ATIS corpus to nine
languages across four language families, and evaluate our method using the
corpus. Results show that our method outperforms a simple label projection
method using fast-align on most languages, and achieves competitive performance
to the more complex, state-of-the-art projection method with only half of the
training time. We release our MultiATIS++ corpus to the community to continue
future research on cross-lingual NLU.
| 2,020 | Computation and Language |
Learning to Learn to Disambiguate: Meta-Learning for Few-Shot Word Sense
Disambiguation | The success of deep learning methods hinges on the availability of large
training datasets annotated for the task of interest. In contrast to human
intelligence, these methods lack versatility and struggle to learn and adapt
quickly to new tasks, where labeled data is scarce. Meta-learning aims to solve
this problem by training a model on a large number of few-shot tasks, with an
objective to learn new tasks quickly from a small number of examples. In this
paper, we propose a meta-learning framework for few-shot word sense
disambiguation (WSD), where the goal is to learn to disambiguate unseen words
from only a few labeled instances. Meta-learning approaches have so far been
typically tested in an $N$-way, $K$-shot classification setting where each task
has $N$ classes with $K$ examples per class. Owing to its nature, WSD deviates
from this controlled setup and requires the models to handle a large number of
highly unbalanced classes. We extend several popular meta-learning approaches
to this scenario, and analyze their strengths and weaknesses in this new
challenging setting.
| 2,020 | Computation and Language |
AxCell: Automatic Extraction of Results from Machine Learning Papers | Tracking progress in machine learning has become increasingly difficult with
the recent explosion in the number of papers. In this paper, we present AxCell,
an automatic machine learning pipeline for extracting results from papers.
AxCell uses several novel components, including a table segmentation subtask,
to learn relevant structural knowledge that aids extraction. When compared with
existing methods, our approach significantly improves the state of the art for
results extraction. We also release a structured, annotated dataset for
training models for results extraction, and a dataset for evaluating the
performance of models on this task. Lastly, we show the viability of our
approach enables it to be used for semi-automated results extraction in
production, suggesting our improvements make this task practically viable for
the first time. Code is available on GitHub.
| 2,020 | Computation and Language |
Detecting Domain Polarity-Changes of Words in a Sentiment Lexicon | Sentiment lexicons are instrumental for sentiment analysis. One can use a set
of sentiment words provided in a sentiment lexicon and a lexicon-based
classifier to perform sentiment classification. One major issue with this
approach is that many sentiment words are domain dependent. That is, they may
be positive in some domains but negative in some others. We refer to this
problem as domain polarity-changes of words. Detecting such words and
correcting their sentiment for an application domain is very important. In this
paper, we propose a graph-based technique to tackle this problem. Experimental
results show its effectiveness on multiple real-world datasets.
| 2,020 | Computation and Language |
Informed Sampling for Diversity in Concept-to-Text NLG | Deep-learning models for language generation tasks tend to produce repetitive
output. Various methods have been proposed to encourage lexical diversity
during decoding, but this often comes at a cost to the perceived fluency and
adequacy of the output. In this work, we propose to ameliorate this cost by
using an Imitation Learning approach to explore the level of diversity that a
language generation model can reliably produce. Specifically, we augment the
decoding process with a meta-classifier trained to distinguish which words at
any given timestep will lead to high-quality output. We focus our experiments
on concept-to-text generation where models are sensitive to the inclusion of
irrelevant words due to the strict relation between input and output. Our
analysis shows that previous methods for diversity underperform in this
setting, while human evaluation suggests that our proposed method achieves a
high level of diversity with minimal effect to the output's fluency and
adequacy.
| 2,021 | Computation and Language |
Elastic weight consolidation for better bias inoculation | The biases present in training datasets have been shown to affect models for
sentence pair classification tasks such as natural language inference (NLI) and
fact verification. While fine-tuning models on additional data has been used to
mitigate them, a common issue is that of catastrophic forgetting of the
original training dataset. In this paper, we show that elastic weight
consolidation (EWC) allows fine-tuning of models to mitigate biases while being
less susceptible to catastrophic forgetting. In our evaluation on fact
verification and NLI stress tests, we show that fine-tuning with EWC dominates
standard fine-tuning, yielding models with lower levels of forgetting on the
original (biased) dataset for equivalent gains in accuracy on the fine-tuning
(unbiased) dataset.
| 2,021 | Computation and Language |
ToTTo: A Controlled Table-To-Text Generation Dataset | 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.
| 2,020 | Computation and Language |
A Benchmark Dataset of Check-worthy Factual Claims | In this paper we present the ClaimBuster dataset of 23,533 statements
extracted from all U.S. general election presidential debates and annotated by
human coders. The ClaimBuster dataset can be leveraged in building
computational methods to identify claims that are worth fact-checking from the
myriad of sources of digital or traditional media. The ClaimBuster dataset is
publicly available to the research community, and it can be found at
http://doi.org/10.5281/zenodo.3609356.
| 2,020 | Computation and Language |
Distantly-Supervised Neural Relation Extraction with Side Information
using BERT | Relation extraction (RE) consists in categorizing the relationship between
entities in a sentence. A recent paradigm to develop relation extractors is
Distant Supervision (DS), which allows the automatic creation of new datasets
by taking an alignment between a text corpus and a Knowledge Base (KB). KBs can
sometimes also provide additional information to the RE task. One of the
methods that adopt this strategy is the RESIDE model, which proposes a
distantly-supervised neural relation extraction using side information from
KBs. Considering that this method outperformed state-of-the-art baselines, in
this paper, we propose a related approach to RESIDE also using additional side
information, but simplifying the sentence encoding with BERT embeddings.
Through experiments, we show the effectiveness of the proposed method in Google
Distant Supervision and Riedel datasets concerning the BGWA and RESIDE baseline
methods. Although Area Under the Curve is decreased because of unbalanced
datasets, P@N results have shown that the use of BERT as sentence encoding
allows superior performance to baseline methods.
| 2,020 | Computation and Language |
What Happens To BERT Embeddings During Fine-tuning? | While there has been much recent work studying how linguistic information is
encoded in pre-trained sentence representations, comparatively little is
understood about how these models change when adapted to solve downstream
tasks. Using a suite of analysis techniques (probing classifiers,
Representational Similarity Analysis, and model ablations), we investigate how
fine-tuning affects the representations of the BERT model. We find that while
fine-tuning necessarily makes significant changes, it does not lead to
catastrophic forgetting of linguistic phenomena. We instead find that
fine-tuning primarily affects the top layers of BERT, but with noteworthy
variation across tasks. In particular, dependency parsing reconfigures most of
the model, whereas SQuAD and MNLI appear to involve much shallower processing.
Finally, we also find that fine-tuning has a weaker effect on representations
of out-of-domain sentences, suggesting room for improvement in model
generalization.
| 2,020 | Computation and Language |
Pragmatic Issue-Sensitive Image Captioning | Image captioning systems have recently improved dramatically, but they still
tend to produce captions that are insensitive to the communicative goals that
captions should meet. To address this, we propose Issue-Sensitive Image
Captioning (ISIC). In ISIC, a captioning system is given a target image and an
issue, which is a set of images partitioned in a way that specifies what
information is relevant. The goal of the captioner is to produce a caption that
resolves this issue. To model this task, we use an extension of the Rational
Speech Acts model of pragmatic language use. Our extension is built on top of
state-of-the-art pretrained neural image captioners and explicitly reasons
about issues in our sense. We establish experimentally that these models
generate captions that are both highly descriptive and issue-sensitive, and we
show how ISIC can complement and enrich the related task of Visual Question
Answering.
| 2,020 | Computation and Language |
SOLID: A Large-Scale Semi-Supervised Dataset for Offensive Language
Identification | The widespread use of offensive content in social media has led to an
abundance of research in detecting language such as hate speech, cyberbullying,
and cyber-aggression. Recent work presented the OLID dataset, which follows a
taxonomy for offensive language identification that provides meaningful
information for understanding the type and the target of offensive messages.
However, it is limited in size and it might be biased towards offensive
language as it was collected using keywords. In this work, we present SOLID, an
expanded dataset, where the tweets were collected in a more principled manner.
SOLID contains over nine million English tweets labeled in a semi-supervised
fashion. We demonstrate that using SOLID along with OLID yields sizable
performance gains on the OLID test set for two different models, especially for
the lower levels of the taxonomy.
| 2,021 | Computation and Language |
"The Boating Store Had Its Best Sail Ever": Pronunciation-attentive
Contextualized Pun Recognition | Humor plays an important role in human languages and it is essential to model
humor when building intelligence systems. Among different forms of humor, puns
perform wordplay for humorous effects by employing words with double entendre
and high phonetic similarity. However, identifying and modeling puns are
challenging as puns usually involved implicit semantic or phonological tricks.
In this paper, we propose Pronunciation-attentive Contextualized Pun
Recognition (PCPR) to perceive human humor, detect if a sentence contains puns
and locate them in the sentence. PCPR derives contextualized representation for
each word in a sentence by capturing the association between the surrounding
context and its corresponding phonetic symbols. Extensive experiments are
conducted on two benchmark datasets. Results demonstrate that the proposed
approach significantly outperforms the state-of-the-art methods in pun
detection and location tasks. In-depth analyses verify the effectiveness and
robustness of PCPR.
| 2,020 | Computation and Language |
Posterior Calibrated Training on Sentence Classification Tasks | Most classification models work by first predicting a posterior probability
distribution over all classes and then selecting that class with the largest
estimated probability. In many settings however, the quality of posterior
probability itself (e.g., 65% chance having diabetes), gives more reliable
information than the final predicted class alone. When these methods are shown
to be poorly calibrated, most fixes to date have relied on posterior
calibration, which rescales the predicted probabilities but often has little
impact on final classifications. Here we propose an end-to-end training
procedure called posterior calibrated (PosCal) training that directly optimizes
the objective while minimizing the difference between the predicted and
empirical posterior probabilities.We show that PosCal not only helps reduce the
calibration error but also improve task performance by penalizing drops in
performance of both objectives. Our PosCal achieves about 2.5% of task
performance gain and 16.1% of calibration error reduction on GLUE (Wang et al.,
2018) compared to the baseline. We achieved the comparable task performance
with 13.2% calibration error reduction on xSLUE (Kang and Hovy, 2019), but not
outperforming the two-stage calibration baseline. PosCal training can be easily
extendable to any types of classification tasks as a form of regularization
term. Also, PosCal has the advantage that it incrementally tracks needed
statistics for the calibration objective during the training process, making
efficient use of large training sets.
| 2,020 | Computation and Language |
Asking without Telling: Exploring Latent Ontologies in Contextual
Representations | The success of pretrained contextual encoders, such as ELMo and BERT, has
brought a great deal of interest in what these models learn: do they, without
explicit supervision, learn to encode meaningful notions of linguistic
structure? If so, how is this structure encoded? To investigate this, we
introduce latent subclass learning (LSL): a modification to existing
classifier-based probing methods that induces a latent categorization (or
ontology) of the probe's inputs. Without access to fine-grained gold labels,
LSL extracts emergent structure from input representations in an interpretable
and quantifiable form. In experiments, we find strong evidence of familiar
categories, such as a notion of personhood in ELMo, as well as novel
ontological distinctions, such as a preference for fine-grained semantic roles
on core arguments. Our results provide unique new evidence of emergent
structure in pretrained encoders, including departures from existing
annotations which are inaccessible to earlier methods.
| 2,020 | Computation and Language |
Instance-Based Learning of Span Representations: A Case Study through
Named Entity Recognition | Interpretable rationales for model predictions play a critical role in
practical applications. In this study, we develop models possessing
interpretable inference process for structured prediction. Specifically, we
present a method of instance-based learning that learns similarities between
spans. At inference time, each span is assigned a class label based on its
similar spans in the training set, where it is easy to understand how much each
training instance contributes to the predictions. Through empirical analysis on
named entity recognition, we demonstrate that our method enables to build
models that have high interpretability without sacrificing performance.
| 2,020 | Computation and Language |
A Supervised Word Alignment Method based on Cross-Language Span
Prediction using Multilingual BERT | We present a novel supervised word alignment method based on cross-language
span prediction. We first formalize a word alignment problem as a collection of
independent predictions from a token in the source sentence to a span in the
target sentence. As this is equivalent to a SQuAD v2.0 style question answering
task, we then solve this problem by using multilingual BERT, which is
fine-tuned on a manually created gold word alignment data. We greatly improved
the word alignment accuracy by adding the context of the token to the question.
In the experiments using five word alignment datasets among Chinese, Japanese,
German, Romanian, French, and English, we show that the proposed method
significantly outperformed previous supervised and unsupervised word alignment
methods without using any bitexts for pretraining. For example, we achieved an
F1 score of 86.7 for the Chinese-English data, which is 13.3 points higher than
the previous state-of-the-art supervised methods.
| 2,020 | Computation and Language |
Bilingual Text Extraction as Reading Comprehension | In this paper, we propose a method to extract bilingual texts automatically
from noisy parallel corpora by framing the problem as a token-level span
prediction, such as SQuAD-style Reading Comprehension. To extract a span of the
target document that is a translation of a given source sentence (span), we use
either QANet or multilingual BERT. QANet can be trained for a specific parallel
corpus from scratch, while multilingual BERT can utilize pre-trained
multilingual representations. For the span prediction method using QANet, we
introduce a total optimization method using integer linear programming to
achieve consistency in the predicted parallel spans. We conduct a parallel
sentence extraction experiment using simulated noisy parallel corpora with two
language pairs (En-Fr and En-Ja) and find that the proposed method using QANet
achieves significantly better accuracy than a baseline method using two
bi-directional RNN encoders, particularly for distant language pairs (En-Ja).
We also conduct a sentence alignment experiment using En-Ja newspaper articles
and find that the proposed method using multilingual BERT achieves
significantly better accuracy than a baseline method using a bilingual
dictionary and dynamic programming.
| 2,020 | Computation and Language |
An Empirical Study of Pre-trained Transformers for Arabic Information
Extraction | Multilingual pre-trained Transformers, such as mBERT (Devlin et al., 2019)
and XLM-RoBERTa (Conneau et al., 2020a), have been shown to enable the
effective cross-lingual zero-shot transfer. However, their performance on
Arabic information extraction (IE) tasks is not very well studied. In this
paper, we pre-train a customized bilingual BERT, dubbed GigaBERT, that is
designed specifically for Arabic NLP and English-to-Arabic zero-shot transfer
learning. We study GigaBERT's effectiveness on zero-short transfer across four
IE tasks: named entity recognition, part-of-speech tagging, argument role
labeling, and relation extraction. Our best model significantly outperforms
mBERT, XLM-RoBERTa, and AraBERT (Antoun et al., 2020) in both the supervised
and zero-shot transfer settings. We have made our pre-trained models publicly
available at https://github.com/lanwuwei/GigaBERT.
| 2,020 | Computation and Language |
Exploiting Sentence Order in Document Alignment | We present a simple document alignment method that incorporates sentence
order information in both candidate generation and candidate re-scoring. Our
method results in 61% relative reduction in error compared to the best
previously published result on the WMT16 document alignment shared task. Our
method improves downstream MT performance on web-scraped Sinhala--English
documents from ParaCrawl, outperforming the document alignment method used in
the most recent ParaCrawl release. It also outperforms a comparable corpora
method which uses the same multilingual embeddings, demonstrating that
exploiting sentence order is beneficial even if the end goal is sentence-level
bitext.
| 2,020 | Computation and Language |
Simulated Multiple Reference Training Improves Low-Resource Machine
Translation | Many valid translations exist for a given sentence, yet machine translation
(MT) is trained with a single reference translation, exacerbating data sparsity
in low-resource settings. We introduce Simulated Multiple Reference Training
(SMRT), a novel MT training method that approximates the full space of possible
translations by sampling a paraphrase of the reference sentence from a
paraphraser and training the MT model to predict the paraphraser's distribution
over possible tokens. We demonstrate the effectiveness of SMRT in low-resource
settings when translating to English, with improvements of 1.2 to 7.0 BLEU. We
also find SMRT is complementary to back-translation.
| 2,021 | Computation and Language |
Stay Hungry, Stay Focused: Generating Informative and Specific Questions
in Information-Seeking Conversations | We investigate the problem of generating informative questions in
information-asymmetric conversations. Unlike previous work on question
generation which largely assumes knowledge of what the answer might be, we are
interested in the scenario where the questioner is not given the context from
which answers are drawn, but must reason pragmatically about how to acquire new
information, given the shared conversation history. We identify two core
challenges: (1) formally defining the informativeness of potential questions,
and (2) exploring the prohibitively large space of potential questions to find
the good candidates. To generate pragmatic questions, we use reinforcement
learning to optimize an informativeness metric we propose, combined with a
reward function designed to promote more specific questions. We demonstrate
that the resulting pragmatic questioner substantially improves the
informativeness and specificity of questions generated over a baseline model,
as evaluated by our metrics as well as humans.
| 2,020 | Computation and Language |
Hierarchical Encoders for Modeling and Interpreting Screenplays | While natural language understanding of long-form documents is still an open
challenge, such documents often contain structural information that can inform
the design of models for encoding them. Movie scripts are an example of such
richly structured text - scripts are segmented into scenes, which are further
decomposed into dialogue and descriptive components. In this work, we propose a
neural architecture for encoding this structure, which performs robustly on a
pair of multi-label tag classification datasets, without the need for
handcrafted features. We add a layer of insight by augmenting an unsupervised
"interpretability" module to the encoder, allowing for the extraction and
visualization of narrative trajectories. Though this work specifically tackles
screenplays, we discuss how the underlying approach can be generalized to a
range of structured documents.
| 2,020 | Computation and Language |
Text Segmentation by Cross Segment Attention | Document and discourse segmentation are two fundamental NLP tasks pertaining
to breaking up text into constituents, which are commonly used to help
downstream tasks such as information retrieval or text summarization. In this
work, we propose three transformer-based architectures and provide
comprehensive comparisons with previously proposed approaches on three standard
datasets. We establish a new state-of-the-art, reducing in particular the error
rates by a large margin in all cases. We further analyze model sizes and find
that we can build models with many fewer parameters while keeping good
performance, thus facilitating real-world applications.
| 2,020 | Computation and Language |
TAVAT: Token-Aware Virtual Adversarial Training for Language
Understanding | Gradient-based adversarial training is widely used in improving the
robustness of neural networks, while it cannot be easily adapted to natural
language processing tasks since the embedding space is discrete. In natural
language processing fields, virtual adversarial training is introduced since
texts are discrete and cannot be perturbed by gradients directly.
Alternatively, virtual adversarial training, which generates perturbations on
the embedding space, is introduced in NLP tasks. Despite its success, existing
virtual adversarial training methods generate perturbations roughly constrained
by Frobenius normalization balls. To craft fine-grained perturbations, we
propose a Token-Aware Virtual Adversarial Training method. We introduce a
token-level accumulated perturbation vocabulary to initialize the perturbations
better and use a token-level normalization ball to constrain these
perturbations pertinently. Experiments show that our method improves the
performance of pre-trained models such as BERT and ALBERT in various tasks by a
considerable margin. The proposed method improves the score of the GLUE
benchmark from 78.3 to 80.9 using BERT model and it also enhances the
performance of sequence labeling and text classification tasks.
| 2,020 | Computation and Language |
WT5?! Training Text-to-Text Models to Explain their Predictions | Neural networks have recently achieved human-level performance on various
challenging natural language processing (NLP) tasks, but it is notoriously
difficult to understand why a neural network produced a particular prediction.
In this paper, we leverage the text-to-text framework proposed by Raffel et
al.(2019) to train language models to output a natural text explanation
alongside their prediction. Crucially, this requires no modifications to the
loss function or training and decoding procedures -- we simply train the model
to output the explanation after generating the (natural text) prediction. We
show that this approach not only obtains state-of-the-art results on
explainability benchmarks, but also permits learning from a limited set of
labeled explanations and transferring rationalization abilities across
datasets. To facilitate reproducibility and future work, we release our code
use to train the models.
| 2,020 | Computation and Language |
Filtering before Iteratively Referring for Knowledge-Grounded Response
Selection in Retrieval-Based Chatbots | The challenges of building knowledge-grounded retrieval-based chatbots lie in
how to ground a conversation on its background knowledge and how to match
response candidates with both context and knowledge simultaneously. This paper
proposes a method named Filtering before Iteratively REferring (FIRE) for this
task. In this method, a context filter and a knowledge filter are first built,
which derive knowledge-aware context representations and context-aware
knowledge representations respectively by global and bidirectional attention.
Besides, the entries irrelevant to the conversation are discarded by the
knowledge filter. After that, iteratively referring is performed between
context and response representations as well as between knowledge and response
representations, in order to collect deep matching features for scoring
response candidates. Experimental results show that FIRE outperforms previous
methods by margins larger than 2.8% and 4.1% on the PERSONA-CHAT dataset with
original and revised personas respectively, and margins larger than 3.1% on the
CMU_DoG dataset in terms of top-1 accuracy. We also show that FIRE is more
interpretable by visualizing the knowledge grounding process.
| 2,020 | Computation and Language |
Indirect Identification of Psychosocial Risks from Natural Language | During the perinatal period, psychosocial health risks, including depression
and intimate partner violence, are associated with serious adverse health
outcomes for parents and children. To appropriately intervene, healthcare
professionals must first identify those at risk, yet stigma often prevents
people from directly disclosing the information needed to prompt an assessment.
We examine indirect methods of eliciting and analyzing information that could
indicate psychosocial risks. Short diary entries by peripartum women exhibit
thematic patterns, extracted by topic modeling, and emotional perspective,
drawn from dictionary-informed sentiment features. Using these features, we use
regularized regression to predict screening measures of depression and
psychological aggression by an intimate partner. Journal text entries
quantified through topic models and sentiment features show promise for
depression prediction, with performance almost as good as closed-form
questions. Text-based features were less useful for prediction of intimate
partner violence, but moderately indirect multiple-choice questioning allowed
for detection without explicit disclosure. Both methods may serve as an initial
or complementary screening approach to detecting stigmatized risks.
| 2,020 | Computation and Language |
User-Guided Aspect Classification for Domain-Specific Texts | Aspect classification, identifying aspects of text segments, facilitates
numerous applications, such as sentiment analysis and review summarization. To
alleviate the human effort on annotating massive texts, in this paper, we study
the problem of classifying aspects based on only a few user-provided seed words
for pre-defined aspects. The major challenge lies in how to handle the noisy
misc aspect, which is designed for texts without any pre-defined aspects. Even
domain experts have difficulties to nominate seed words for the misc aspect,
making existing seed-driven text classification methods not applicable. We
propose a novel framework, ARYA, which enables mutual enhancements between
pre-defined aspects and the misc aspect via iterative classifier training and
seed updating. Specifically, it trains a classifier for pre-defined aspects and
then leverages it to induce the supervision for the misc aspect. The prediction
results of the misc aspect are later utilized to filter out noisy seed words
for pre-defined aspects. Experiments in two domains demonstrate the superior
performance of our proposed framework, as well as the necessity and importance
of properly modeling the misc aspect.
| 2,020 | Computation and Language |
RikiNet: Reading Wikipedia Pages for Natural Question Answering | Reading long documents to answer open-domain questions remains challenging in
natural language understanding. In this paper, we introduce a new model, called
RikiNet, which reads Wikipedia pages for natural question answering. RikiNet
contains a dynamic paragraph dual-attention reader and a multi-level cascaded
answer predictor. The reader dynamically represents the document and question
by utilizing a set of complementary attention mechanisms. The representations
are then fed into the predictor to obtain the span of the short answer, the
paragraph of the long answer, and the answer type in a cascaded manner. On the
Natural Questions (NQ) dataset, a single RikiNet achieves 74.3 F1 and 57.9 F1
on long-answer and short-answer tasks. To our best knowledge, it is the first
single model that outperforms the single human performance. Furthermore, an
ensemble RikiNet obtains 76.1 F1 and 61.3 F1 on long-answer and short-answer
tasks, achieving the best performance on the official NQ leaderboard
| 2,020 | Computation and Language |
Automatic Machine Translation Evaluation in Many Languages via Zero-Shot
Paraphrasing | We frame the task of machine translation evaluation as one of scoring machine
translation output with a sequence-to-sequence paraphraser, conditioned on a
human reference. We propose training the paraphraser as a multilingual NMT
system, treating paraphrasing as a zero-shot translation task (e.g., Czech to
Czech). This results in the paraphraser's output mode being centered around a
copy of the input sequence, which represents the best case scenario where the
MT system output matches a human reference. Our method is simple and intuitive,
and does not require human judgements for training. Our single model (trained
in 39 languages) outperforms or statistically ties with all prior metrics on
the WMT 2019 segment-level shared metrics task in all languages (excluding
Gujarati where the model had no training data). We also explore using our model
for the task of quality estimation as a metric--conditioning on the source
instead of the reference--and find that it significantly outperforms every
submission to the WMT 2019 shared task on quality estimation in every language
pair.
| 2,020 | Computation and Language |
Boosting Naturalness of Language in Task-oriented Dialogues via
Adversarial Training | The natural language generation (NLG) module in a task-oriented dialogue
system produces user-facing utterances conveying required information. Thus, it
is critical for the generated response to be natural and fluent. We propose to
integrate adversarial training to produce more human-like responses. The model
uses Straight-Through Gumbel-Softmax estimator for gradient computation. We
also propose a two-stage training scheme to boost performance. Empirical
results show that the adversarial training can effectively improve the quality
of language generation in both automatic and human evaluations. For example, in
the RNN-LG Restaurant dataset, our model AdvNLG outperforms the previous
state-of-the-art result by 3.6% in BLEU.
| 2,020 | Computation and Language |
memeBot: Towards Automatic Image Meme Generation | Image memes have become a widespread tool used by people for interacting and
exchanging ideas over social media, blogs, and open messengers. This work
proposes to treat automatic image meme generation as a translation process, and
further present an end to end neural and probabilistic approach to generate an
image-based meme for any given sentence using an encoder-decoder architecture.
For a given input sentence, an image meme is generated by combining a meme
template image and a text caption where the meme template image is selected
from a set of popular candidates using a selection module, and the meme caption
is generated by an encoder-decoder model. An encoder is used to map the
selected meme template and the input sentence into a meme embedding and a
decoder is used to decode the meme caption from the meme embedding. The
generated natural language meme caption is conditioned on the input sentence
and the selected meme template. The model learns the dependencies between the
meme captions and the meme template images and generates new memes using the
learned dependencies. The quality of the generated captions and the generated
memes is evaluated through both automated and human evaluation. An experiment
is designed to score how well the generated memes can represent the tweets from
Twitter conversations. Experiments on Twitter data show the efficacy of the
model in generating memes for sentences in online social interaction.
| 2,020 | Computation and Language |
Exploring Contextualized Neural Language Models for Temporal Dependency
Parsing | Extracting temporal relations between events and time expressions has many
applications such as constructing event timelines and time-related question
answering. It is a challenging problem which requires syntactic and semantic
information at sentence or discourse levels, which may be captured by deep
contextualized language models (LMs) such as BERT (Devlin et al., 2019). In
this paper, we develop several variants of BERT-based temporal dependency
parser, and show that BERT significantly improves temporal dependency parsing
(Zhang and Xue, 2018a). We also present a detailed analysis on why deep
contextualized neural LMs help and where they may fall short. Source code and
resources are made available at https://github.com/bnmin/tdp_ranking.
| 2,020 | Computation and Language |
Logic2Text: High-Fidelity Natural Language Generation from Logical Forms | Previous works on Natural Language Generation (NLG) from structured data have
primarily focused on surface-level descriptions of record sequences. However,
for complex structured data, e.g., multi-row tables, it is often desirable for
an NLG system to describe interesting facts from logical inferences across
records. If only provided with the table, it is hard for existing models to
produce controllable and high-fidelity logical generations. In this work, we
formulate logical level NLG as generation from logical forms in order to obtain
controllable, high-fidelity, and faithful generations. We present a new
large-scale dataset, \textsc{Logic2Text}, with 10,753 descriptions involving
common logic types paired with the underlying logical forms. The logical forms
show diversified graph structure of free schema, which poses great challenges
on the model's ability to understand the semantics. We experiment on (1)
Fully-supervised training with the full datasets, and (2) Few-shot setting,
provided with hundreds of paired examples; We compare several popular
generation models and analyze their performances. We hope our dataset can
encourage research towards building an advanced NLG system capable of natural,
faithful, and human-like generation. The dataset and code are available at
https://github.com/czyssrs/Logic2Text.
| 2,020 | Computation and Language |
Improved Natural Language Generation via Loss Truncation | Neural language models are usually trained to match the distributional
properties of a large-scale corpus by minimizing the log loss. While
straightforward to optimize, this approach forces the model to reproduce all
variations in the dataset, including noisy and invalid references (e.g.,
misannotation and hallucinated facts). Worse, the commonly used log loss is
overly sensitive to such phenomena and even a small fraction of noisy data can
degrade performance. In this work, we show that the distinguishability of the
models and reference serves as a principled and robust alternative for handling
invalid references. To optimize distinguishability, we propose loss truncation,
which adaptively removes high loss examples during training. We show this is as
easy to optimize as log loss and tightly bounds distinguishability under noise.
Empirically, we demonstrate that loss truncation outperforms existing baselines
on distinguishability on a summarization task, and show that samples generated
by the loss truncation model have factual accuracy ratings that exceed those of
baselines and match human references.
| 2,020 | Computation and Language |
EnsembleGAN: Adversarial Learning for Retrieval-Generation Ensemble
Model on Short-Text Conversation | Generating qualitative responses has always been a challenge for
human-computer dialogue systems. Existing dialogue systems generally derive
from either retrieval-based or generative-based approaches, both of which have
their own pros and cons. Despite the natural idea of an ensemble model of the
two, existing ensemble methods only focused on leveraging one approach to
enhance another, we argue however that they can be further mutually enhanced
with a proper training strategy. In this paper, we propose ensembleGAN, an
adversarial learning framework for enhancing a retrieval-generation ensemble
model in open-domain conversation scenario. It consists of a
language-model-like generator, a ranker generator, and one ranker
discriminator. Aiming at generating responses that approximate the ground-truth
and receive high ranking scores from the discriminator, the two generators
learn to generate improved highly relevant responses and competitive unobserved
candidates respectively, while the discriminative ranker is trained to identify
true responses from adversarial ones, thus featuring the merits of both
generator counterparts. The experimental results on a large short-text
conversation data demonstrate the effectiveness of the ensembleGAN by the
amelioration on both human and automatic evaluation metrics.
| 2,020 | Computation and Language |
Learning Music Helps You Read: Using Transfer to Study Linguistic
Structure in Language Models | We propose transfer learning as a method for analyzing the encoding of
grammatical structure in neural language models. We train LSTMs on
non-linguistic data and evaluate their performance on natural language to
assess which kinds of data induce generalizable structural features that LSTMs
can use for natural language. We find that training on non-linguistic data with
latent structure (MIDI music or Java code) improves test performance on natural
language, despite no overlap in surface form or vocabulary. To pinpoint the
kinds of abstract structure that models may be encoding to lead to this
improvement, we run similar experiments with two artificial parentheses
languages: one which has a hierarchical recursive structure, and a control
which has paired tokens but no recursion. Surprisingly, training a model on
either of these artificial languages leads to the same substantial gains when
testing on natural language. Further experiments on transfer between natural
languages controlling for vocabulary overlap show that zero-shot performance on
a test language is highly correlated with typological syntactic similarity to
the training language, suggesting that representations induced by pre-training
correspond to the cross-linguistic syntactic properties. Our results provide
insights into the ways that neural models represent abstract syntactic
structure, and also about the kind of structural inductive biases which allow
for natural language acquisition.
| 2,020 | Computation and Language |
Look at the First Sentence: Position Bias in Question Answering | Many extractive question answering models are trained to predict start and
end positions of answers. The choice of predicting answers as positions is
mainly due to its simplicity and effectiveness. In this study, we hypothesize
that when the distribution of the answer positions is highly skewed in the
training set (e.g., answers lie only in the k-th sentence of each passage), QA
models predicting answers as positions can learn spurious positional cues and
fail to give answers in different positions. We first illustrate this position
bias in popular extractive QA models such as BiDAF and BERT and thoroughly
examine how position bias propagates through each layer of BERT. To safely
deliver position information without position bias, we train models with
various de-biasing methods including entropy regularization and bias
ensembling. Among them, we found that using the prior distribution of answer
positions as a bias model is very effective at reducing position bias,
recovering the performance of BERT from 37.48% to 81.64% when trained on a
biased SQuAD dataset.
| 2,021 | Computation and Language |
Can Your Context-Aware MT System Pass the DiP Benchmark Tests? :
Evaluation Benchmarks for Discourse Phenomena in Machine Translation | Despite increasing instances of machine translation (MT) systems including
contextual information, the evidence for translation quality improvement is
sparse, especially for discourse phenomena. Popular metrics like BLEU are not
expressive or sensitive enough to capture quality improvements or drops that
are minor in size but significant in perception. We introduce the first of
their kind MT benchmark datasets that aim to track and hail improvements across
four main discourse phenomena: anaphora, lexical consistency, coherence and
readability, and discourse connective translation. We also introduce evaluation
methods for these tasks, and evaluate several baseline MT systems on the
curated datasets. Surprisingly, we find that existing context-aware models do
not improve discourse-related translations consistently across languages and
phenomena.
| 2,020 | Computation and Language |
Knowledge Injection into Dialogue Generation via Language Models | Dialogue generation has been successfully learned from scratch by neural
networks, but tends to produce the same general response, e.g., "what are you
talking about?", in many conversations. To reduce this homogeneity, external
knowledge such as the speaker's profile and domain knowledge is applied as an
additional condition to diversify a model's output. The required knowledge to
develop an effective conversation, however, is not always available, which is
different from prior work's assumption that a model always has acquired
sufficient knowledge before chatting. This problem can be detrimental when
applying a dialogue model like this chatting online with unconstrained people
and topics, because the model does not have the needed knowledge. To address
this problem, we propose InjK, which is a two-stage approach to inject
knowledge into a dialogue generation model. First, we train a large-scale
language model and query it as textual knowledge. Second, we frame a dialogue
generation model to sequentially generate textual knowledge and a corresponding
response. Empirically, when a dialogue generation model can only access limited
knowledge, our method outperforms prior work by producing more coherent and
informative responses.
| 2,021 | Computation and Language |
Universal Dependencies according to BERT: both more specific and more
general | This work focuses on analyzing the form and extent of syntactic abstraction
captured by BERT by extracting labeled dependency trees from self-attentions.
Previous work showed that individual BERT heads tend to encode particular
dependency relation types. We extend these findings by explicitly comparing
BERT relations to Universal Dependencies (UD) annotations, showing that they
often do not match one-to-one.
We suggest a method for relation identification and syntactic tree
construction. Our approach produces significantly more consistent dependency
trees than previous work, showing that it better explains the syntactic
abstractions in BERT. At the same time, it can be successfully applied with
only a minimal amount of supervision and generalizes well across languages.
| 2,020 | Computation and Language |
Neural Natural Language Inference Models Partially Embed Theories of
Lexical Entailment and Negation | We address whether neural models for Natural Language Inference (NLI) can
learn the compositional interactions between lexical entailment and negation,
using four methods: the behavioral evaluation methods of (1) challenge test
sets and (2) systematic generalization tasks, and the structural evaluation
methods of (3) probes and (4) interventions. To facilitate this holistic
evaluation, we present Monotonicity NLI (MoNLI), a new naturalistic dataset
focused on lexical entailment and negation. In our behavioral evaluations, we
find that models trained on general-purpose NLI datasets fail systematically on
MoNLI examples containing negation, but that MoNLI fine-tuning addresses this
failure. In our structural evaluations, we look for evidence that our
top-performing BERT-based model has learned to implement the monotonicity
algorithm behind MoNLI. Probes yield evidence consistent with this conclusion,
and our intervention experiments bolster this, showing that the causal dynamics
of the model mirror the causal dynamics of this algorithm on subsets of MoNLI.
This suggests that the BERT model at least partially embeds a theory of lexical
entailment and negation at an algorithmic level.
| 2,020 | Computation and Language |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.