Titles
stringlengths 6
220
| Abstracts
stringlengths 37
3.26k
| Years
int64 1.99k
2.02k
| Categories
stringclasses 1
value |
|---|---|---|---|
Key Phrase Extraction & Applause Prediction
|
With the increase in content availability over the internet it is very
difficult to get noticed. It has become an upmost the priority of the blog
writers to get some feedback over their creations to be confident about the
impact of their article. We are training a machine learning model to learn
popular article styles, in the form of vector space representations using
various word embeddings, and their popularity based on claps and tags.
| 2,021 |
Computation and Language
|
SDA: Improving Text Generation with Self Data Augmentation
|
Data augmentation has been widely used to improve deep neural networks in
many research fields, such as computer vision. However, less work has been done
in the context of text, partially due to its discrete nature and the complexity
of natural languages. In this paper, we propose to improve the standard maximum
likelihood estimation (MLE) paradigm by incorporating a self-imitation-learning
phase for automatic data augmentation. Unlike most existing sentence-level
augmentation strategies, which are only applied to specific models, our method
is more general and could be easily adapted to any MLE-based training
procedure. In addition, our framework allows task-specific evaluation metrics
to be designed to flexibly control the generated sentences, for example, in
terms of controlling vocabulary usage and avoiding nontrivial repetitions.
Extensive experimental results demonstrate the superiority of our method on two
synthetic and several standard real datasets, significantly improving related
baselines.
| 2,021 |
Computation and Language
|
Learning to Emphasize: Dataset and Shared Task Models for Selecting
Emphasis in Presentation Slides
|
Presentation slides have become a common addition to the teaching material.
Emphasizing strong leading words in presentation slides can allow the audience
to direct the eye to certain focal points instead of reading the entire slide,
retaining the attention to the speaker during the presentation. Despite a large
volume of studies on automatic slide generation, few studies have addressed the
automation of design assistance during the creation process. Motivated by this
demand, we study the problem of Emphasis Selection (ES) in presentation slides,
i.e., choosing candidates for emphasis, by introducing a new dataset containing
presentation slides with a wide variety of topics, each is annotated with
emphasis words in a crowdsourced setting. We evaluate a range of
state-of-the-art models on this novel dataset by organizing a shared task and
inviting multiple researchers to model emphasis in this new domain. We present
the main findings and compare the results of these models, and by examining the
challenges of the dataset, we provide different analysis components.
| 2,021 |
Computation and Language
|
A Gamification of Japanese Dependency Parsing
|
Gamification approaches have been used as a way for creating language
resources for NLP. It is also used for presenting and teaching the algorithms
in NLP and linguistic phenomena. This paper argues about a design of
gamification for Japanese syntactic dependendency parsing for the latter
objective. The user interface design is based on a transition-based shift
reduce dependency parsing which needs only two actions of SHIFT (not attach)
and REDUCE (attach) in Japanese dependency structure. We assign the two actions
for two-way directional control on a gamepad or other devices. We also design
the target sentences from psycholinguistics researches.
| 2,021 |
Computation and Language
|
Unifying Relational Sentence Generation and Retrieval for Medical Image
Report Composition
|
Beyond generating long and topic-coherent paragraphs in traditional
captioning tasks, the medical image report composition task poses more
task-oriented challenges by requiring both the highly-accurate medical term
diagnosis and multiple heterogeneous forms of information including impression
and findings. Current methods often generate the most common sentences due to
dataset bias for individual case, regardless of whether the sentences properly
capture key entities and relationships. Such limitations severely hinder their
applicability and generalization capability in medical report composition where
the most critical sentences lie in the descriptions of abnormal diseases that
are relatively rare. Moreover, some medical terms appearing in one report are
often entangled with each other and co-occurred, e.g. symptoms associated with
a specific disease. To enforce the semantic consistency of medical terms to be
incorporated into the final reports and encourage the sentence generation for
rare abnormal descriptions, we propose a novel framework that unifies template
retrieval and sentence generation to handle both common and rare abnormality
while ensuring the semantic-coherency among the detected medical terms.
Specifically, our approach exploits hybrid-knowledge co-reasoning: i) explicit
relationships among all abnormal medical terms to induce the visual attention
learning and topic representation encoding for better topic-oriented symptoms
descriptions; ii) adaptive generation mode that changes between the template
retrieval and sentence generation according to a contextual topic encoder.
Experimental results on two medical report benchmarks demonstrate the
superiority of the proposed framework in terms of both human and metrics
evaluation.
| 2,021 |
Computation and Language
|
Trankit: A Light-Weight Transformer-based Toolkit for Multilingual
Natural Language Processing
|
We introduce Trankit, a light-weight Transformer-based Toolkit for
multilingual Natural Language Processing (NLP). It provides a trainable
pipeline for fundamental NLP tasks over 100 languages, and 90 pretrained
pipelines for 56 languages. Built on a state-of-the-art pretrained language
model, Trankit significantly outperforms prior multilingual NLP pipelines over
sentence segmentation, part-of-speech tagging, morphological feature tagging,
and dependency parsing while maintaining competitive performance for
tokenization, multi-word token expansion, and lemmatization over 90 Universal
Dependencies treebanks. Despite the use of a large pretrained transformer, our
toolkit is still efficient in memory usage and speed. This is achieved by our
novel plug-and-play mechanism with Adapters where a multilingual pretrained
transformer is shared across pipelines for different languages. Our toolkit
along with pretrained models and code are publicly available at:
https://github.com/nlp-uoregon/trankit. A demo website for our toolkit is also
available at: http://nlp.uoregon.edu/trankit. Finally, we create a demo video
for Trankit at: https://youtu.be/q0KGP3zGjGc.
| 2,021 |
Computation and Language
|
Combating Hostility: Covid-19 Fake News and Hostile Post Detection in
Social Media
|
This paper illustrates a detail description of the system and its results
that developed as a part of the participation at CONSTRAINT shared task in
AAAI-2021. The shared task comprises two tasks: a) COVID19 fake news detection
in English b) Hostile post detection in Hindi. Task-A is a binary
classification problem with fake and real class, while task-B is a multi-label
multi-class classification task with five hostile classes (i.e. defame, fake,
hate, offense, non-hostile). Various techniques are used to perform the
classification task, including SVM, CNN, BiLSTM, and CNN+BiLSTM with tf-idf and
Word2Vec embedding techniques. Results indicate that SVM with tf-idf features
achieved the highest 94.39% weighted $f_1$ score on the test set in task-A.
Label powerset SVM with n-gram features obtained the maximum coarse-grained and
fine-grained $f_1$ score of 86.03% and 50.98% on the task-B test set
respectively.
| 2,021 |
Computation and Language
|
LightXML: Transformer with Dynamic Negative Sampling for
High-Performance Extreme Multi-label Text Classification
|
Extreme Multi-label text Classification (XMC) is a task of finding the most
relevant labels from a large label set. Nowadays deep learning-based methods
have shown significant success in XMC. However, the existing methods (e.g.,
AttentionXML and X-Transformer etc) still suffer from 1) combining several
models to train and predict for one dataset, and 2) sampling negative labels
statically during the process of training label ranking model, which reduces
both the efficiency and accuracy of the model. To address the above problems,
we proposed LightXML, which adopts end-to-end training and dynamic negative
labels sampling. In LightXML, we use generative cooperative networks to recall
and rank labels, in which label recalling part generates negative and positive
labels, and label ranking part distinguishes positive labels from these labels.
Through these networks, negative labels are sampled dynamically during label
ranking part training by feeding with the same text representation. Extensive
experiments show that LightXML outperforms state-of-the-art methods in five
extreme multi-label datasets with much smaller model size and lower
computational complexity. In particular, on the Amazon dataset with 670K
labels, LightXML can reduce the model size up to 72% compared to AttentionXML.
| 2,021 |
Computation and Language
|
Learning Better Sentence Representation with Syntax Information
|
Sentence semantic understanding is a key topic in the field of natural
language processing. Recently, contextualized word representations derived from
pre-trained language models such as ELMO and BERT have shown significant
improvements for a wide range of semantic tasks, e.g. question answering, text
classification and sentiment analysis. However, how to add external knowledge
to further improve the semantic modeling capability of model is worth probing.
In this paper, we propose a novel approach to combining syntax information with
a pre-trained language model. In order to evaluate the effect of the
pre-training model, first, we introduce RNN-based and Transformer-based
pre-trained language models; secondly, to better integrate external knowledge,
such as syntactic information integrate with the pre-training model, we propose
a dependency syntax expansion (DSE) model. For evaluation, we have selected two
subtasks: sentence completion task and biological relation extraction task. The
experimental results show that our model achieves 91.2\% accuracy,
outperforming the baseline model by 37.8\% on sentence completion task. And it
also gets competitive performance by 75.1\% $F_{1}$ score on relation
extraction task.
| 2,021 |
Computation and Language
|
Task Adaptive Pretraining of Transformers for Hostility Detection
|
Identifying adverse and hostile content on the web and more particularly, on
social media, has become a problem of paramount interest in recent years. With
their ever increasing popularity, fine-tuning of pretrained Transformer-based
encoder models with a classifier head are gradually becoming the new baseline
for natural language classification tasks. In our work, we explore the gains
attributed to Task Adaptive Pretraining (TAPT) prior to fine-tuning of
Transformer-based architectures. We specifically study two problems, namely,
(a) Coarse binary classification of Hindi Tweets into Hostile or Not, and (b)
Fine-grained multi-label classification of Tweets into four categories: hate,
fake, offensive, and defamation. Building up on an architecture which takes
emojis and segmented hashtags into consideration for classification, we are
able to experimentally showcase the performance upgrades due to TAPT. Our
system (with team name 'iREL IIIT') ranked first in the 'Hostile Post Detection
in Hindi' shared task with an F1 score of 97.16% for coarse-grained detection
and a weighted F1 score of 62.96% for fine-grained multi-label classification
on the provided blind test corpora.
| 2,021 |
Computation and Language
|
BERT & Family Eat Word Salad: Experiments with Text Understanding
|
In this paper, we study the response of large models from the BERT family to
incoherent inputs that should confuse any model that claims to understand
natural language. We define simple heuristics to construct such examples. Our
experiments show that state-of-the-art models consistently fail to recognize
them as ill-formed, and instead produce high confidence predictions on them. As
a consequence of this phenomenon, models trained on sentences with randomly
permuted word order perform close to state-of-the-art models. To alleviate
these issues, we show that if models are explicitly trained to recognize
invalid inputs, they can be robust to such attacks without a drop in
performance.
| 2,021 |
Computation and Language
|
Detecting Hostile Posts using Relational Graph Convolutional Network
|
This work is based on the submission to the competition Hindi Constraint
conducted by AAAI@2021 for detection of hostile posts in Hindi on social media
platforms. Here, a model is presented for detection and classification of
hostile posts and further classify into fake, offensive, hate and defamation
using Relational Graph Convolutional Networks. Unlike other existing work, our
approach is focused on using semantic meaning along with contextutal
information for better classification. The results from AAAI@2021 indicates
that the proposed model is performing at par with Google's XLM-RoBERTa on the
given dataset. Our best submission with RGCN achieves an F1 score of 0.97 (7th
Rank) on coarse-grained evaluation and achieved best performance on identifying
fake posts. Among all submissions to the challenge, our classification system
with XLM-Roberta secured 2nd rank on fine-grained classification.
| 2,021 |
Computation and Language
|
Adaptive Prototypical Networks with Label Words and Joint Representation
Learning for Few-Shot Relation Classification
|
Relation classification (RC) task is one of fundamental tasks of information
extraction, aiming to detect the relation information between entity pairs in
unstructured natural language text and generate structured data in the form of
entity-relation triple. Although distant supervision methods can effectively
alleviate the problem of lack of training data in supervised learning, they
also introduce noise into the data, and still cannot fundamentally solve the
long-tail distribution problem of the training instances. In order to enable
the neural network to learn new knowledge through few instances like humans,
this work focuses on few-shot relation classification (FSRC), where a
classifier should generalize to new classes that have not been seen in the
training set, given only a number of samples for each class. To make full use
of the existing information and get a better feature representation for each
instance, we propose to encode each class prototype in an adaptive way from two
aspects. First, based on the prototypical networks, we propose an adaptive
mixture mechanism to add label words to the representation of the class
prototype, which, to the best of our knowledge, is the first attempt to
integrate the label information into features of the support samples of each
class so as to get more interactive class prototypes. Second, to more
reasonably measure the distances between samples of each category, we introduce
a loss function for joint representation learning to encode each support
instance in an adaptive manner. Extensive experiments have been conducted on
FewRel under different few-shot (FS) settings, and the results show that the
proposed adaptive prototypical networks with label words and joint
representation learning has not only achieved significant improvements in
accuracy, but also increased the generalization ability of few-shot RC models.
| 2,021 |
Computation and Language
|
A Heuristic-driven Ensemble Framework for COVID-19 Fake News Detection
|
The significance of social media has increased manifold in the past few
decades as it helps people from even the most remote corners of the world stay
connected. With the COVID-19 pandemic raging, social media has become more
relevant and widely used than ever before, and along with this, there has been
a resurgence in the circulation of fake news and tweets that demand immediate
attention. In this paper, we describe our Fake News Detection system that
automatically identifies whether a tweet related to COVID-19 is "real" or
"fake", as a part of CONSTRAINT COVID19 Fake News Detection in English
challenge. We have used an ensemble model consisting of pre-trained models that
has helped us achieve a joint 8th position on the leader board. We have
achieved an F1-score of 0.9831 against a top score of 0.9869. Post completion
of the competition, we have been able to drastically improve our system by
incorporating a novel heuristic algorithm based on username handles and link
domains in tweets fetching an F1-score of 0.9883 and achieving state-of-the art
results on the given dataset.
| 2,021 |
Computation and Language
|
Summaformers @ LaySumm 20, LongSumm 20
|
Automatic text summarization has been widely studied as an important task in
natural language processing. Traditionally, various feature engineering and
machine learning based systems have been proposed for extractive as well as
abstractive text summarization. Recently, deep learning based, specifically
Transformer-based systems have been immensely popular. Summarization is a
cognitively challenging task - extracting summary worthy sentences is
laborious, and expressing semantics in brief when doing abstractive
summarization is complicated. In this paper, we specifically look at the
problem of summarizing scientific research papers from multiple domains. We
differentiate between two types of summaries, namely, (a) LaySumm: A very short
summary that captures the essence of the research paper in layman terms
restricting overtly specific technical jargon and (b) LongSumm: A much longer
detailed summary aimed at providing specific insights into various ideas
touched upon in the paper. While leveraging latest Transformer-based models,
our systems are simple, intuitive and based on how specific paper sections
contribute to human summaries of the two types described above. Evaluations
against gold standard summaries using ROUGE metrics prove the effectiveness of
our approach. On blind test corpora, our system ranks first and third for the
LongSumm and LaySumm tasks respectively.
| 2,020 |
Computation and Language
|
The Logic for a Mildly Context-Sensitive Fragment of the Lambek-Grishin
Calculus
|
While context-free grammars are characterized by a simple proof-theoretic
grammatical formalism namely categorial grammar and its logic the Lambek
calculus, no such characterizations were known for tree-adjoining grammars, and
even for any mildly context-sensitive languages classes in the last forty years
despite some efforts. We settle this problem in this paper. On the basis of the
existing fragment of the Lambek-Grishin calculus which captures tree-adjoining
languages, we present a logic called HLG: a proof-theoretic characterization of
tree-adjoining languages based on the Lambek-Grishin calculus restricted to
Hyperedge-replacement grammar with rank two studied by Moot. HLG is defined in
display calculus with cut-admissibility. Several new techniques are introduced
for the proofs, such as purely structural connectives, usefulness, and a
graph-theoretic argument on proof nets for HLG.
| 2,021 |
Computation and Language
|
AT-BERT: Adversarial Training BERT for Acronym Identification Winning
Solution for SDU@AAAI-21
|
Acronym identification focuses on finding the acronyms and the phrases that
have been abbreviated, which is crucial for scientific document understanding
tasks. However, the limited size of manually annotated datasets hinders further
improvement for the problem. Recent breakthroughs of language models
pre-trained on large corpora clearly show that unsupervised pre-training can
vastly improve the performance of downstream tasks. In this paper, we present
an Adversarial Training BERT method named AT-BERT, our winning solution to
acronym identification task for Scientific Document Understanding (SDU)
Challenge of AAAI 2021. Specifically, the pre-trained BERT is adopted to
capture better semantic representation. Then we incorporate the FGM adversarial
training strategy into the fine-tuning of BERT, which makes the model more
robust and generalized. Furthermore, an ensemble mechanism is devised to
involve the representations learned from multiple BERT variants. Assembling all
these components together, the experimental results on the SciAI dataset show
that our proposed approach outperforms all other competitive state-of-the-art
methods.
| 2,021 |
Computation and Language
|
Constraint 2021: Machine Learning Models for COVID-19 Fake News
Detection Shared Task
|
In this system paper we present our contribution to the Constraint 2021
COVID-19 Fake News Detection Shared Task, which poses the challenge of
classifying COVID-19 related social media posts as either fake or real. In our
system, we address this challenge by applying classical machine learning
algorithms together with several linguistic features, such as n-grams,
readability, emotional tone and punctuation. In terms of pre-processing, we
experiment with various steps like stop word removal, stemming/lemmatization,
link removal and more. We find our best performing system to be based on a
linear SVM, which obtains a weighted average F1 score of 95.19% on test data,
which lands a place in the middle of the leaderboard (place 80 of 167).
| 2,021 |
Computation and Language
|
Improving Multi-hop Knowledge Base Question Answering by Learning
Intermediate Supervision Signals
|
Multi-hop Knowledge Base Question Answering (KBQA) aims to find the answer
entities that are multiple hops away in the Knowledge Base (KB) from the
entities in the question. A major challenge is the lack of supervision signals
at intermediate steps. Therefore, multi-hop KBQA algorithms can only receive
the feedback from the final answer, which makes the learning unstable or
ineffective.
To address this challenge, we propose a novel teacher-student approach for
the multi-hop KBQA task. In our approach, the student network aims to find the
correct answer to the query, while the teacher network tries to learn
intermediate supervision signals for improving the reasoning capacity of the
student network. The major novelty lies in the design of the teacher network,
where we utilize both forward and backward reasoning to enhance the learning of
intermediate entity distributions. By considering bidirectional reasoning, the
teacher network can produce more reliable intermediate supervision signals,
which can alleviate the issue of spurious reasoning. Extensive experiments on
three benchmark datasets have demonstrated the effectiveness of our approach on
the KBQA task. The code to reproduce our analysis is available at
https://github.com/RichardHGL/WSDM2021_NSM.
| 2,021 |
Computation and Language
|
Revisiting Mahalanobis Distance for Transformer-Based Out-of-Domain
Detection
|
Real-life applications, heavily relying on machine learning, such as dialog
systems, demand out-of-domain detection methods. Intent classification models
should be equipped with a mechanism to distinguish seen intents from unseen
ones so that the dialog agent is capable of rejecting the latter and avoiding
undesired behavior. However, despite increasing attention paid to the task, the
best practices for out-of-domain intent detection have not yet been fully
established.
This paper conducts a thorough comparison of out-of-domain intent detection
methods. We prioritize the methods, not requiring access to out-of-domain data
during training, gathering of which is extremely time- and labor-consuming due
to lexical and stylistic variation of user utterances. We evaluate multiple
contextual encoders and methods, proven to be efficient, on three standard
datasets for intent classification, expanded with out-of-domain utterances. Our
main findings show that fine-tuning Transformer-based encoders on in-domain
data leads to superior results. Mahalanobis distance, together with utterance
representations, derived from Transformer-based encoders, outperforms other
methods by a wide margin and establishes new state-of-the-art results for all
datasets.
The broader analysis shows that the reason for success lies in the fact that
the fine-tuned Transformer is capable of constructing homogeneous
representations of in-domain utterances, revealing geometrical disparity to out
of domain utterances. In turn, the Mahalanobis distance captures this disparity
easily.
The code is available in our GitHub repo:
https://github.com/huawei-noah/noah-research/tree/master/Maha_OOD .
| 2,022 |
Computation and Language
|
Model Generalization on COVID-19 Fake News Detection
|
Amid the pandemic COVID-19, the world is facing unprecedented infodemic with
the proliferation of both fake and real information. Considering the
problematic consequences that the COVID-19 fake-news have brought, the
scientific community has put effort to tackle it. To contribute to this fight
against the infodemic, we aim to achieve a robust model for the COVID-19
fake-news detection task proposed at CONSTRAINT 2021 (FakeNews-19) by taking
two separate approaches: 1) fine-tuning transformers based language models with
robust loss functions and 2) removing harmful training instances through
influence calculation. We further evaluate the robustness of our models by
evaluating on different COVID-19 misinformation test set (Tweets-19) to
understand model generalization ability. With the first approach, we achieve
98.13% for weighted F1 score (W-F1) for the shared task, whereas 38.18% W-F1 on
the Tweets-19 highest. On the contrary, by performing influence data cleansing,
our model with 99% cleansing percentage can achieve 54.33% W-F1 score on
Tweets-19 with a trade-off. By evaluating our models on two COVID-19 fake-news
test sets, we suggest the importance of model generalization ability in this
task to step forward to tackle the COVID-19 fake-news problem in online social
media platforms.
| 2,021 |
Computation and Language
|
edATLAS: An Efficient Disambiguation Algorithm for Texting in Languages
with Abugida Scripts
|
Abugida refers to a phonogram writing system where each syllable is
represented using a single consonant or typographic ligature, along with a
default vowel or optional diacritic(s) to denote other vowels. However, texting
in these languages has some unique challenges in spite of the advent of devices
with soft keyboard supporting custom key layouts. The number of characters in
these languages is large enough to require characters to be spread over
multiple views in the layout. Having to switch between views many times to type
a single word hinders the natural thought process. This prevents popular usage
of native keyboard layouts. On the other hand, supporting romanized scripts
(native words transcribed using Latin characters) with language model based
suggestions is also set back by the lack of uniform romanization rules.
To this end, we propose a disambiguation algorithm and showcase its
usefulness in two novel mutually non-exclusive input methods for languages
natively using the abugida writing system: (a) disambiguation of ambiguous
input for abugida scripts, and (b) disambiguation of word variants in romanized
scripts. We benchmark these approaches using public datasets, and show an
improvement in typing speed by 19.49%, 25.13%, and 14.89%, in Hindi, Bengali,
and Thai, respectively, using Ambiguous Input, owing to the human ease of
locating keys combined with the efficiency of our inference method. Our Word
Variant Disambiguation (WDA) maps valid variants of romanized words, previously
treated as Out-of-Vocab, to a vocabulary of 100k words with high accuracy,
leading to an increase in Error Correction F1 score by 10.03% and Next Word
Prediction (NWP) by 62.50% on average.
| 2,021 |
Computation and Language
|
Language Detection Engine for Multilingual Texting on Mobile Devices
|
More than 2 billion mobile users worldwide type in multiple languages in the
soft keyboard. On a monolingual keyboard, 38% of falsely auto-corrected words
are valid in another language. This can be easily avoided by detecting the
language of typed words and then validating it in its respective language.
Language detection is a well-known problem in natural language processing. In
this paper, we present a fast, light-weight and accurate Language Detection
Engine (LDE) for multilingual typing that dynamically adapts to user intended
language in real-time. We propose a novel approach where the fusion of
character N-gram model and logistic regression based selector model is used to
identify the language. Additionally, we present a unique method of reducing the
inference time significantly by parameter reduction technique. We also discuss
various optimizations fabricated across LDE to resolve ambiguity in input text
among the languages with the same character pattern. Our method demonstrates an
average accuracy of 94.5% for Indian languages in Latin script and that of 98%
for European languages on the code-switched data. This model outperforms
fastText by 60.39% and ML-Kit by 23.67% in F1 score for European languages. LDE
is faster on mobile device with an average inference time of 25.91
microseconds.
| 2,020 |
Computation and Language
|
Real-Time Optimized N-gram For Mobile Devices
|
With the increasing number of mobile devices, there has been continuous
research on generating optimized Language Models (LMs) for soft keyboard. In
spite of advances in this domain, building a single LM for low-end feature
phones as well as high-end smartphones is still a pressing need. Hence, we
propose a novel technique, Optimized N-gram (Op-Ngram), an end-to-end N-gram
pipeline that utilises mobile resources efficiently for faster Word Completion
(WC) and Next Word Prediction (NWP). Op-Ngram applies Stupid Backoff and
pruning strategies to generate a light-weight model. The LM loading time on
mobile is linear with respect to model size. We observed that Op-Ngram gives
37% improvement in Language Model (LM)-ROM size, 76% in LM-RAM size, 88% in
loading time and 89% in average suggestion time as compared to SORTED array
variant of BerkeleyLM. Moreover, our method shows significant performance
improvement over KenLM as well.
| 2,019 |
Computation and Language
|
Identification of COVID-19 related Fake News via Neural Stacking
|
Identification of Fake News plays a prominent role in the ongoing pandemic,
impacting multiple aspects of day-to-day life. In this work we present a
solution to the shared task titled COVID19 Fake News Detection in English,
scoring the 50th place amongst 168 submissions. The solution was within 1.5% of
the best performing solution. The proposed solution employs a heterogeneous
representation ensemble, adapted for the classification task via an additional
neural classification head comprised of multiple hidden layers. The paper
consists of detailed ablation studies further displaying the proposed method's
behavior and possible implications. The solution is freely available.
\url{https://gitlab.com/boshko.koloski/covid19-fake-news}
| 2,021 |
Computation and Language
|
Context- and Sequence-Aware Convolutional Recurrent Encoder for Neural
Machine Translation
|
Neural Machine Translation model is a sequence-to-sequence converter based on
neural networks. Existing models use recurrent neural networks to construct
both the encoder and decoder modules. In alternative research, the recurrent
networks were substituted by convolutional neural networks for capturing the
syntactic structure in the input sentence and decreasing the processing time.
We incorporate the goodness of both approaches by proposing a
convolutional-recurrent encoder for capturing the context information as well
as the sequential information from the source sentence. Word embedding and
position embedding of the source sentence is performed prior to the
convolutional encoding layer which is basically a n-gram feature extractor
capturing phrase-level context information. The rectified output of the
convolutional encoding layer is added to the original embedding vector, and the
sum is normalized by layer normalization. The normalized output is given as a
sequential input to the recurrent encoding layer that captures the temporal
information in the sequence. For the decoder, we use the attention-based
recurrent neural network. Translation task on the German-English dataset
verifies the efficacy of the proposed approach from the higher BLEU scores
achieved as compared to the state of the art.
| 2,021 |
Computation and Language
|
Automating the Compilation of Potential Core-Outcomes for Clinical
Trials
|
Due to increased access to clinical trial outcomes and analysis, researchers
and scientists are able to iterate or improve upon relevant approaches more
effectively. However, the metrics and related results of clinical trials
typically do not follow any standardization in their reports, making it more
difficult for researchers to parse the results of different trials. The
objective of this paper is to describe an automated method utilizing natural
language processing in order to describe the probable core outcomes of clinical
trials, in order to alleviate the issues around disparate clinical trial
outcomes. As the nature of this process is domain specific, BioBERT was
employed in order to conduct a multi-class entity normalization task. In
addition to BioBERT, an unsupervised feature-based approach making use of only
the encoder output embedding representations for the outcomes and labels was
utilized. Finally, cosine similarity was calculated across the vectors to
obtain the semantic similarity. This method was able to both harness the
domain-specific context of each of the tokens from the learned embeddings of
the BioBERT model as well as a more stable metric of sentence similarity. Some
common outcomes identified using the Jaccard similarity in each of the
classifications were compiled, and while some are untenable, a pipeline for
which this automation process could be conducted was established.
| 2,021 |
Computation and Language
|
Explain and Predict, and then Predict Again
|
A desirable property of learning systems is to be both effective and
interpretable. Towards this goal, recent models have been proposed that first
generate an extractive explanation from the input text and then generate a
prediction on just the explanation called explain-then-predict models. These
models primarily consider the task input as a supervision signal in learning an
extractive explanation and do not effectively integrate rationales data as an
additional inductive bias to improve task performance. We propose a novel yet
simple approach ExPred, that uses multi-task learning in the explanation
generation phase effectively trading-off explanation and prediction losses. And
then we use another prediction network on just the extracted explanations for
optimizing the task performance. We conduct an extensive evaluation of our
approach on three diverse language datasets -- fact verification, sentiment
classification, and QA -- and find that we substantially outperform existing
approaches.
| 2,021 |
Computation and Language
|
Evaluation of Deep Learning Models for Hostility Detection in Hindi Text
|
The social media platform is a convenient medium to express personal thoughts
and share useful information. It is fast, concise, and has the ability to reach
millions. It is an effective place to archive thoughts, share artistic content,
receive feedback, promote products, etc. Despite having numerous advantages
these platforms have given a boost to hostile posts. Hate speech and derogatory
remarks are being posted for personal satisfaction or political gain. The
hostile posts can have a bullying effect rendering the entire platform
experience hostile. Therefore detection of hostile posts is important to
maintain social media hygiene. The problem is more pronounced languages like
Hindi which are low in resources. In this work, we present approaches for
hostile text detection in the Hindi language. The proposed approaches are
evaluated on the Constraint@AAAI 2021 Hindi hostility detection dataset. The
dataset consists of hostile and non-hostile texts collected from social media
platforms. The hostile posts are further segregated into overlapping classes of
fake, offensive, hate, and defamation. We evaluate a host of deep learning
approaches based on CNN, LSTM, and BERT for this multi-label classification
problem. The pre-trained Hindi fast text word embeddings by IndicNLP and
Facebook are used in conjunction with CNN and LSTM models. Two variations of
pre-trained multilingual transformer language models mBERT and IndicBERT are
used. We show that the performance of BERT based models is best. Moreover, CNN
and LSTM models also perform competitively with BERT based models.
| 2,021 |
Computation and Language
|
BERT-GT: Cross-sentence n-ary relation extraction with BERT and Graph
Transformer
|
A biomedical relation statement is commonly expressed in multiple sentences
and consists of many concepts, including gene, disease, chemical, and mutation.
To automatically extract information from biomedical literature, existing
biomedical text-mining approaches typically formulate the problem as a
cross-sentence n-ary relation-extraction task that detects relations among n
entities across multiple sentences, and use either a graph neural network (GNN)
with long short-term memory (LSTM) or an attention mechanism. Recently,
Transformer has been shown to outperform LSTM on many natural language
processing (NLP) tasks. In this work, we propose a novel architecture that
combines Bidirectional Encoder Representations from Transformers with Graph
Transformer (BERT-GT), through integrating a neighbor-attention mechanism into
the BERT architecture. Unlike the original Transformer architecture, which
utilizes the whole sentence(s) to calculate the attention of the current token,
the neighbor-attention mechanism in our method calculates its attention
utilizing only its neighbor tokens. Thus, each token can pay attention to its
neighbor information with little noise. We show that this is critically
important when the text is very long, as in cross-sentence or abstract-level
relation-extraction tasks. Our benchmarking results show improvements of 5.44%
and 3.89% in accuracy and F1-measure over the state-of-the-art on n-ary and
chemical-protein relation datasets, suggesting BERT-GT is a robust approach
that is applicable to other biomedical relation extraction tasks or datasets.
| 2,021 |
Computation and Language
|
Clustering Word Embeddings with Self-Organizing Maps. Application on
LaRoSeDa -- A Large Romanian Sentiment Data Set
|
Romanian is one of the understudied languages in computational linguistics,
with few resources available for the development of natural language processing
tools. In this paper, we introduce LaRoSeDa, a Large Romanian Sentiment Data
Set, which is composed of 15,000 positive and negative reviews collected from
one of the largest Romanian e-commerce platforms. We employ two sentiment
classification methods as baselines for our new data set, one based on
low-level features (character n-grams) and one based on high-level features
(bag-of-word-embeddings generated by clustering word embeddings with k-means).
As an additional contribution, we replace the k-means clustering algorithm with
self-organizing maps (SOMs), obtaining better results because the generated
clusters of word embeddings are closer to the Zipf's law distribution, which is
known to govern natural language. We also demonstrate the generalization
capacity of using SOMs for the clustering of word embeddings on another
recently-introduced Romanian data set, for text categorization by topic.
| 2,021 |
Computation and Language
|
Implicit Unlikelihood Training: Improving Neural Text Generation with
Reinforcement Learning
|
Likelihood training and maximization-based decoding result in dull and
repetitive generated texts even when using powerful language models (Holtzman
et al., 2019). Adding a loss function for regularization was shown to improve
text generation output by helping avoid unwanted properties, such as
contradiction or repetition (Li at al., 2020). In this work, we propose
fine-tuning a language model by using policy gradient reinforcement learning,
directly optimizing for better generation. We apply this approach to minimizing
repetition in generated text, and show that, when combined with unlikelihood
training (Welleck et al., 2020), our method further reduces repetition without
impacting the language model quality. We also evaluate other methods for
improving generation at training and decoding time, and compare them using
various metrics aimed at control for better text generation output.
| 2,021 |
Computation and Language
|
Transforming Multi-Conditioned Generation from Meaning Representation
|
In task-oriented conversation systems, natural language generation systems
that generate sentences with specific information related to conversation flow
are useful. Our study focuses on language generation by considering various
information representing the meaning of utterances as multiple conditions of
generation. NLG from meaning representations, the conditions for sentence
meaning, generally goes through two steps: sentence planning and surface
realization. However, we propose a simple one-stage framework to generate
utterances directly from MR (Meaning Representation). Our model is based on
GPT2 and generates utterances with flat conditions on slot and value pairs,
which does not need to determine the structure of the sentence. We evaluate
several systems in the E2E dataset with 6 automatic metrics. Our system is a
simple method, but it demonstrates comparable performance to previous systems
in automated metrics. In addition, using only 10\% of the data set without any
other techniques, our model achieves comparable performance, and shows the
possibility of performing zero-shot generation and expanding to other datasets.
| 2,021 |
Computation and Language
|
Neural Contract Element Extraction Revisited: Letters from Sesame Street
|
We investigate contract element extraction. We show that LSTM-based encoders
perform better than dilated CNNs, Transformers, and BERT in this task. We also
find that domain-specific WORD2VEC embeddings outperform generic pre-trained
GLOVE embeddings. Morpho-syntactic features in the form of POS tag and token
shape embeddings, as well as context-aware ELMO embeddings do not improve
performance. Several of these observations contradict choices or findings of
previous work on contract element extraction and generic sequence labeling
tasks, indicating that contract element extraction requires careful
task-specific choices. Analyzing the results of (i) plain TRANSFORMER-based and
(ii) BERT-based models, we find that in the examined task, where the entities
are highly context-sensitive, the lack of recurrency in TRANSFORMERs greatly
affects their performance.
| 2,019 |
Computation and Language
|
Quantum Cognitively Motivated Decision Fusion for Video Sentiment
Analysis
|
Video sentiment analysis as a decision-making process is inherently complex,
involving the fusion of decisions from multiple modalities and the so-caused
cognitive biases. Inspired by recent advances in quantum cognition, we show
that the sentiment judgment from one modality could be incompatible with the
judgment from another, i.e., the order matters and they cannot be jointly
measured to produce a final decision. Thus the cognitive process exhibits
"quantum-like" biases that cannot be captured by classical probability
theories. Accordingly, we propose a fundamentally new, quantum cognitively
motivated fusion strategy for predicting sentiment judgments. In particular, we
formulate utterances as quantum superposition states of positive and negative
sentiment judgments, and uni-modal classifiers as mutually incompatible
observables, on a complex-valued Hilbert space with positive-operator valued
measures. Experiments on two benchmarking datasets illustrate that our model
significantly outperforms various existing decision level and a range of
state-of-the-art content-level fusion approaches. The results also show that
the concept of incompatibility allows effective handling of all combination
patterns, including those extreme cases that are wrongly predicted by all
uni-modal classifiers.
| 2,021 |
Computation and Language
|
A character representation enhanced on-device Intent Classification
|
Intent classification is an important task in natural language understanding
systems. Existing approaches have achieved perfect scores on the benchmark
datasets. However they are not suitable for deployment on low-resource devices
like mobiles, tablets, etc. due to their massive model size. Therefore, in this
paper, we present a novel light-weight architecture for intent classification
that can run efficiently on a device. We use character features to enrich the
word representation. Our experiments prove that our proposed model outperforms
existing approaches and achieves state-of-the-art results on benchmark
datasets. We also report that our model has tiny memory footprint of ~5 MB and
low inference time of ~2 milliseconds, which proves its efficiency in a
resource-constrained environment.
| 2,021 |
Computation and Language
|
Of Non-Linearity and Commutativity in BERT
|
In this work we provide new insights into the transformer architecture, and
in particular, its best-known variant, BERT. First, we propose a method to
measure the degree of non-linearity of different elements of transformers.
Next, we focus our investigation on the feed-forward networks (FFN) inside
transformers, which contain 2/3 of the model parameters and have so far not
received much attention. We find that FFNs are an inefficient yet important
architectural element and that they cannot simply be replaced by attention
blocks without a degradation in performance. Moreover, we study the
interactions between layers in BERT and show that, while the layers exhibit
some hierarchical structure, they extract features in a fuzzy manner. Our
results suggest that BERT has an inductive bias towards layer commutativity,
which we find is mainly due to the skip connections. This provides a
justification for the strong performance of recurrent and weight-shared
transformer models.
| 2,021 |
Computation and Language
|
Toward Effective Automated Content Analysis via Crowdsourcing
|
Many computer scientists use the aggregated answers of online workers to
represent ground truth. Prior work has shown that aggregation methods such as
majority voting are effective for measuring relatively objective features. For
subjective features such as semantic connotation, online workers, known for
optimizing their hourly earnings, tend to deteriorate in the quality of their
responses as they work longer. In this paper, we aim to address this issue by
proposing a quality-aware semantic data annotation system. We observe that with
timely feedback on workers' performance quantified by quality scores, better
informed online workers can maintain the quality of their labeling throughout
an extended period of time. We validate the effectiveness of the proposed
annotation system through i) evaluating performance based on an expert-labeled
dataset, and ii) demonstrating machine learning tasks that can lead to
consistent learning behavior with 70%-80% accuracy. Our results suggest that
with our system, researchers can collect high-quality answers of subjective
semantic features at a large scale.
| 2,021 |
Computation and Language
|
AI- and HPC-enabled Lead Generation for SARS-CoV-2: Models and Processes
to Extract Druglike Molecules Contained in Natural Language Text
|
Researchers worldwide are seeking to repurpose existing drugs or discover new
drugs to counter the disease caused by severe acute respiratory syndrome
coronavirus 2 (SARS-CoV-2). A promising source of candidates for such studies
is molecules that have been reported in the scientific literature to be
drug-like in the context of coronavirus research. We report here on a project
that leverages both human and artificial intelligence to detect references to
drug-like molecules in free text. We engage non-expert humans to create a
corpus of labeled text, use this labeled corpus to train a named entity
recognition model, and employ the trained model to extract 10912 drug-like
molecules from the COVID-19 Open Research Dataset Challenge (CORD-19) corpus of
198875 papers. Performance analyses show that our automated extraction model
can achieve performance on par with that of non-expert humans.
| 2,021 |
Computation and Language
|
Latent Alignment of Procedural Concepts in Multimodal Recipes
|
We propose a novel alignment mechanism to deal with procedural reasoning on a
newly released multimodal QA dataset, named RecipeQA. Our model is solving the
textual cloze task which is a reading comprehension on a recipe containing
images and instructions. We exploit the power of attention networks,
cross-modal representations, and a latent alignment space between instructions
and candidate answers to solve the problem. We introduce constrained
max-pooling which refines the max-pooling operation on the alignment matrix to
impose disjoint constraints among the outputs of the model. Our evaluation
result indicates a 19\% improvement over the baselines.
| 2,020 |
Computation and Language
|
Self-Training Pre-Trained Language Models for Zero- and Few-Shot
Multi-Dialectal Arabic Sequence Labeling
|
A sufficient amount of annotated data is usually required to fine-tune
pre-trained language models for downstream tasks. Unfortunately, attaining
labeled data can be costly, especially for multiple language varieties and
dialects. We propose to self-train pre-trained language models in zero- and
few-shot scenarios to improve performance on data-scarce varieties using only
resources from data-rich ones. We demonstrate the utility of our approach in
the context of Arabic sequence labeling by using a language model fine-tuned on
Modern Standard Arabic (MSA) only to predict named entities (NE) and
part-of-speech (POS) tags on several dialectal Arabic (DA) varieties. We show
that self-training is indeed powerful, improving zero-shot MSA-to-DA transfer
by as large as \texttildelow 10\% F$_1$ (NER) and 2\% accuracy (POS tagging).
We acquire even better performance in few-shot scenarios with limited amounts
of labeled data. We conduct an ablation study and show that the performance
boost observed directly results from the unlabeled DA examples used for
self-training. Our work opens up opportunities for developing DA models
exploiting only MSA resources and it can be extended to other languages and
tasks. Our code and fine-tuned models can be accessed at
https://github.com/mohammadKhalifa/zero-shot-arabic-dialects.
| 2,021 |
Computation and Language
|
Robustness Gym: Unifying the NLP Evaluation Landscape
|
Despite impressive performance on standard benchmarks, deep neural networks
are often brittle when deployed in real-world systems. Consequently, recent
research has focused on testing the robustness of such models, resulting in a
diverse set of evaluation methodologies ranging from adversarial attacks to
rule-based data transformations. In this work, we identify challenges with
evaluating NLP systems and propose a solution in the form of Robustness Gym
(RG), a simple and extensible evaluation toolkit that unifies 4 standard
evaluation paradigms: subpopulations, transformations, evaluation sets, and
adversarial attacks. By providing a common platform for evaluation, Robustness
Gym enables practitioners to compare results from all 4 evaluation paradigms
with just a few clicks, and to easily develop and share novel evaluation
methods using a built-in set of abstractions. To validate Robustness Gym's
utility to practitioners, we conducted a real-world case study with a
sentiment-modeling team, revealing performance degradations of 18%+. To verify
that Robustness Gym can aid novel research analyses, we perform the first study
of state-of-the-art commercial and academic named entity linking (NEL) systems,
as well as a fine-grained analysis of state-of-the-art summarization models.
For NEL, commercial systems struggle to link rare entities and lag their
academic counterparts by 10%+, while state-of-the-art summarization models
struggle on examples that require abstraction and distillation, degrading by
9%+. Robustness Gym can be found at https://robustnessgym.com/
| 2,021 |
Computation and Language
|
Experimental Evaluation of Deep Learning models for Marathi Text
Classification
|
The Marathi language is one of the prominent languages used in India. It is
predominantly spoken by the people of Maharashtra. Over the past decade, the
usage of language on online platforms has tremendously increased. However,
research on Natural Language Processing (NLP) approaches for Marathi text has
not received much attention. Marathi is a morphologically rich language and
uses a variant of the Devanagari script in the written form. This works aims to
provide a comprehensive overview of available resources and models for Marathi
text classification. We evaluate CNN, LSTM, ULMFiT, and BERT based models on
two publicly available Marathi text classification datasets and present a
comparative analysis. The pre-trained Marathi fast text word embeddings by
Facebook and IndicNLP are used in conjunction with word-based models. We show
that basic single layer models based on CNN and LSTM coupled with FastText
embeddings perform on par with the BERT based models on the available datasets.
We hope our paper aids focused research and experiments in the area of Marathi
NLP.
| 2,022 |
Computation and Language
|
EventPlus: A Temporal Event Understanding Pipeline
|
We present EventPlus, a temporal event understanding pipeline that integrates
various state-of-the-art event understanding components including event trigger
and type detection, event argument detection, event duration and temporal
relation extraction. Event information, especially event temporal knowledge, is
a type of common sense knowledge that helps people understand how stories
evolve and provides predictive hints for future events. EventPlus as the first
comprehensive temporal event understanding pipeline provides a convenient tool
for users to quickly obtain annotations about events and their temporal
information for any user-provided document. Furthermore, we show EventPlus can
be easily adapted to other domains (e.g., biomedical domain). We make EventPlus
publicly available to facilitate event-related information extraction and
downstream applications.
| 2,021 |
Computation and Language
|
LaDiff ULMFiT: A Layer Differentiated training approach for ULMFiT
|
In our paper, we present Deep Learning models with a layer differentiated
training method which were used for the SHARED TASK@ CONSTRAINT 2021 sub-tasks
COVID19 Fake News Detection in English and Hostile Post Detection in Hindi. We
propose a Layer Differentiated training procedure for training a pre-trained
ULMFiT arXiv:1801.06146 model. We used special tokens to annotate specific
parts of the tweets to improve language understanding and gain insights on the
model making the tweets more interpretable. The other two submissions included
a modified RoBERTa model and a simple Random Forest Classifier. The proposed
approach scored a precision and f1 score of 0.96728972 and 0.967324832
respectively for sub-task "COVID19 Fake News Detection in English". Also,
Coarse-Grained Hostility f1 Score and Weighted FineGrained f1 score of 0.908648
and 0.533907 respectively for sub-task Hostile Post Detection in Hindi. The
proposed approach ranked 61st out of 164 in the sub-task "COVID19 Fake News
Detection in English and 18th out of 45 in the sub-task Hostile Post Detection
in Hindi".
| 2,021 |
Computation and Language
|
Improving Commonsense Causal Reasoning by Adversarial Training and Data
Augmentation
|
Determining the plausibility of causal relations between clauses is a
commonsense reasoning task that requires complex inference ability. The general
approach to this task is to train a large pretrained language model on a
specific dataset. However, the available training data for the task is often
scarce, which leads to instability of model training or reliance on the shallow
features of the dataset. This paper presents a number of techniques for making
models more robust in the domain of causal reasoning. Firstly, we perform
adversarial training by generating perturbed inputs through synonym
substitution. Secondly, based on a linguistic theory of discourse connectives,
we perform data augmentation using a discourse parser for detecting causally
linked clauses in large text, and a generative language model for generating
distractors. Both methods boost model performance on the Choice of Plausible
Alternatives (COPA) dataset, as well as on a Balanced COPA dataset, which is a
modified version of the original data that has been developed to avoid
superficial cues, leading to a more challenging benchmark. We show a
statistically significant improvement in performance and robustness on both
datasets, even with only a small number of additionally generated data points.
| 2,021 |
Computation and Language
|
Coarse and Fine-Grained Hostility Detection in Hindi Posts using Fine
Tuned Multilingual Embeddings
|
Due to the wide adoption of social media platforms like Facebook, Twitter,
etc., there is an emerging need of detecting online posts that can go against
the community acceptance standards. The hostility detection task has been well
explored for resource-rich languages like English, but is unexplored for
resource-constrained languages like Hindidue to the unavailability of large
suitable data. We view this hostility detection as a multi-label multi-class
classification problem. We propose an effective neural network-based technique
for hostility detection in Hindi posts. We leverage pre-trained multilingual
Bidirectional Encoder Representations of Transformer (mBERT) to obtain the
contextual representations of Hindi posts. We have performed extensive
experiments including different pre-processing techniques, pre-trained models,
neural architectures, hybrid strategies, etc. Our best performing neural
classifier model includes One-vs-the-Rest approach where we obtained 92.60%,
81.14%,69.59%, 75.29% and 73.01% F1 scores for hostile, fake, hate, offensive,
and defamation labels respectively. The proposed model outperformed the
existing baseline models and emerged as the state-of-the-art model for
detecting hostility in the Hindi posts.
| 2,021 |
Computation and Language
|
Is the User Enjoying the Conversation? A Case Study on the Impact on the
Reward Function
|
The impact of user satisfaction in policy learning task-oriented dialogue
systems has long been a subject of research interest. Most current models for
estimating the user satisfaction either (i) treat out-of-context short-texts,
such as product reviews, or (ii) rely on turn features instead of on
distributed semantic representations. In this work we adopt deep neural
networks that use distributed semantic representation learning for estimating
the user satisfaction in conversations. We evaluate the impact of modelling
context length in these networks. Moreover, we show that the proposed
hierarchical network outperforms state-of-the-art quality estimators.
Furthermore, we show that applying these networks to infer the reward function
in a Partial Observable Markov Decision Process (POMDP) yields to a great
improvement in the task success rate.
| 2,021 |
Computation and Language
|
Uzbek Cyrillic-Latin-Cyrillic Machine Transliteration
|
In this paper, we introduce a data-driven approach to transliterating Uzbek
dictionary words from the Cyrillic script into the Latin script, and vice
versa. We heuristically align characters of words in the source script with
sub-strings of the corresponding words in the target script and train a
decision tree classifier that learns these alignments. On the test set, our
Cyrillic to Latin model achieves a character level micro-averaged F1 score of
0.9992, and our Latin to Cyrillic model achieves the score of 0.9959. Our
contribution is a novel method of producing machine transliterated texts for
the low-resource Uzbek language.
| 2,021 |
Computation and Language
|
On consistency scores in text data with an implementation in R
|
In this paper, we introduce a reproducible cleaning process for the text
extracted from PDFs using n-gram models. Our approach compares the originally
extracted text with the text generated from, or expected by, these models using
earlier text as stimulus. To guide this process, we introduce the notion of a
consistency score, which refers to the proportion of text that is expected by
the model. This is used to monitor changes during the cleaning process, and
across different corpuses. We illustrate our process on text from the book Jane
Eyre and introduce both a Shiny application and an R package to make our
process easier for others to adopt.
| 2,021 |
Computation and Language
|
Machine-Assisted Script Curation
|
We describe Machine-Aided Script Curator (MASC), a system for human-machine
collaborative script authoring. Scripts produced with MASC include (1) English
descriptions of sub-events that comprise a larger, complex event; (2) event
types for each of those events; (3) a record of entities expected to
participate in multiple sub-events; and (4) temporal sequencing between the
sub-events. MASC automates portions of the script creation process with
suggestions for event types, links to Wikidata, and sub-events that may have
been forgotten. We illustrate how these automations are useful to the script
writer with a few case-study scripts.
| 2,021 |
Computation and Language
|
Text Augmentation in a Multi-Task View
|
Traditional data augmentation aims to increase the coverage of the input
distribution by generating augmented examples that strongly resemble original
samples in an online fashion where augmented examples dominate training. In
this paper, we propose an alternative perspective -- a multi-task view (MTV) of
data augmentation -- in which the primary task trains on original examples and
the auxiliary task trains on augmented examples. In MTV data augmentation, both
original and augmented samples are weighted substantively during training,
relaxing the constraint that augmented examples must resemble original data and
thereby allowing us to apply stronger levels of augmentation. In empirical
experiments using four common data augmentation techniques on three benchmark
text classification datasets, we find that the MTV leads to higher and more
robust performance improvements than traditional augmentation.
| 2,021 |
Computation and Language
|
WER-BERT: Automatic WER Estimation with BERT in a Balanced Ordinal
Classification Paradigm
|
Automatic Speech Recognition (ASR) systems are evaluated using Word Error
Rate (WER), which is calculated by comparing the number of errors between the
ground truth and the transcription of the ASR system. This calculation,
however, requires manual transcription of the speech signal to obtain the
ground truth. Since transcribing audio signals is a costly process, Automatic
WER Evaluation (e-WER) methods have been developed to automatically predict the
WER of a speech system by only relying on the transcription and the speech
signal features. While WER is a continuous variable, previous works have shown
that positing e-WER as a classification problem is more effective than
regression. However, while converting to a classification setting, these
approaches suffer from heavy class imbalance. In this paper, we propose a new
balanced paradigm for e-WER in a classification setting. Within this paradigm,
we also propose WER-BERT, a BERT based architecture with speech features for
e-WER. Furthermore, we introduce a distance loss function to tackle the ordinal
nature of e-WER classification. The proposed approach and paradigm are
evaluated on the Librispeech dataset and a commercial (black box) ASR system,
Google Cloud's Speech-to-Text API. The results and experiments demonstrate that
WER-BERT establishes a new state-of-the-art in automatic WER estimation.
| 2,021 |
Computation and Language
|
Hostility Detection in Hindi leveraging Pre-Trained Language Models
|
Hostile content on social platforms is ever increasing. This has led to the
need for proper detection of hostile posts so that appropriate action can be
taken to tackle them. Though a lot of work has been done recently in the
English Language to solve the problem of hostile content online, similar works
in Indian Languages are quite hard to find. This paper presents a transfer
learning based approach to classify social media (i.e Twitter, Facebook, etc.)
posts in Hindi Devanagari script as Hostile or Non-Hostile. Hostile posts are
further analyzed to determine if they are Hateful, Fake, Defamation, and
Offensive. This paper harnesses attention based pre-trained models fine-tuned
on Hindi data with Hostile-Non hostile task as Auxiliary and fusing its
features for further sub-tasks classification. Through this approach, we
establish a robust and consistent model without any ensembling or complex
pre-processing. We have presented the results from our approach in
CONSTRAINT-2021 Shared Task on hostile post detection where our model performs
extremely well with 3rd runner up in terms of Weighted Fine-Grained F1 Score.
| 2,021 |
Computation and Language
|
ECOL: Early Detection of COVID Lies Using Content, Prior Knowledge and
Source Information
|
Social media platforms are vulnerable to fake news dissemination, which
causes negative consequences such as panic and wrong medication in the
healthcare domain. Therefore, it is important to automatically detect fake news
in an early stage before they get widely spread. This paper analyzes the impact
of incorporating content information, prior knowledge, and credibility of
sources into models for the early detection of fake news. We propose a
framework modeling those features by using BERT language model and external
sources, namely Simple English Wikipedia and source reliability tags. The
conducted experiments on CONSTRAINT datasets demonstrated the benefit of
integrating these features for the early detection of fake news in the
healthcare domain.
| 2,021 |
Computation and Language
|
Transformer-based Language Model Fine-tuning Methods for COVID-19 Fake
News Detection
|
With the pandemic of COVID-19, relevant fake news is spreading all over the
sky throughout the social media. Believing in them without discrimination can
cause great trouble to people's life. However, universal language models may
perform weakly in these fake news detection for lack of large-scale annotated
data and sufficient semantic understanding of domain-specific knowledge. While
the model trained on corresponding corpora is also mediocre for insufficient
learning. In this paper, we propose a novel transformer-based language model
fine-tuning approach for these fake news detection. First, the token vocabulary
of individual model is expanded for the actual semantics of professional
phrases. Second, we adapt the heated-up softmax loss to distinguish the
hard-mining samples, which are common for fake news because of the
disambiguation of short text. Then, we involve adversarial training to improve
the model's robustness. Last, the predicted features extracted by universal
language model RoBERTa and domain-specific model CT-BERT are fused by one
multiple layer perception to integrate fine-grained and high-level specific
representations. Quantitative experimental results evaluated on existing
COVID-19 fake news dataset show its superior performances compared to the
state-of-the-art methods among various evaluation metrics. Furthermore, the
best weighted average F1 score achieves 99.02%.
| 2,023 |
Computation and Language
|
On the Temporality of Priors in Entity Linking
|
Entity linking is a fundamental task in natural language processing which
deals with the lexical ambiguity in texts. An important component in entity
linking approaches is the mention-to-entity prior probability. Even though
there is a large number of works in entity linking, the existing approaches do
not explicitly consider the time aspect, specifically the temporality of an
entity's prior probability. We posit that this prior probability is temporal in
nature and affects the performance of entity linking systems. In this paper we
systematically study the effect of the prior on the entity linking performance
over the temporal validity of both texts and KBs.
| 2,020 |
Computation and Language
|
Better Together -- An Ensemble Learner for Combining the Results of
Ready-made Entity Linking Systems
|
Entity linking (EL) is the task of automatically identifying entity mentions
in text and resolving them to a corresponding entity in a reference knowledge
base like Wikipedia. Throughout the past decade, a plethora of EL systems and
pipelines have become available, where performance of individual systems varies
heavily across corpora, languages or domains. Linking performance varies even
between different mentions in the same text corpus, where, for instance, some
EL approaches are better able to deal with short surface forms while others may
perform better when more context information is available. To this end, we
argue that performance may be optimised by exploiting results from distinct EL
systems on the same corpus, thereby leveraging their individual strengths on a
per-mention basis. In this paper, we introduce a supervised approach which
exploits the output of multiple ready-made EL systems by predicting the correct
link on a per-mention basis. Experimental results obtained on existing ground
truth datasets and exploiting three state-of-the-art EL systems show the
effectiveness of our approach and its capacity to significantly outperform the
individual EL systems as well as a set of baseline methods.
| 2,021 |
Computation and Language
|
On Informative Tweet Identification For Tracking Mass Events
|
Twitter has been heavily used as an important channel for communicating and
discussing about events in real-time. In such major events, many uninformative
tweets are also published rapidly by many users, making it hard to follow the
events. In this paper, we address this problem by investigating machine
learning methods for automatically identifying informative tweets among those
that are relevant to a target event. We examine both traditional approaches
with a rich set of handcrafted features and state of the art approaches with
automatically learned features. We further propose a hybrid model that
leverages both the handcrafted features and the automatically learned ones. Our
experiments on several large datasets of real-world events show that the latter
approaches significantly outperform the former and our proposed model performs
the best, suggesting highly effective mechanisms for tracking mass events.
| 2,021 |
Computation and Language
|
TUDublin team at Constraint@AAAI2021 -- COVID19 Fake News Detection
|
The paper is devoted to the participation of the TUDublin team in
Constraint@AAAI2021 - COVID19 Fake News Detection Challenge. Today, the problem
of fake news detection is more acute than ever in connection with the pandemic.
The number of fake news is increasing rapidly and it is necessary to create AI
tools that allow us to identify and prevent the spread of false information
about COVID-19 urgently. The main goal of the work was to create a model that
would carry out a binary classification of messages from social media as real
or fake news in the context of COVID-19. Our team constructed the ensemble
consisting of Bidirectional Long Short Term Memory, Support Vector Machine,
Logistic Regression, Naive Bayes and a combination of Logistic Regression and
Naive Bayes. The model allowed us to achieve 0.94 F1-score, which is within 5\%
of the best result.
| 2,021 |
Computation and Language
|
SICKNL: A Dataset for Dutch Natural Language Inference
|
We present SICK-NL (read: signal), a dataset targeting Natural Language
Inference in Dutch. SICK-NL is obtained by translating the SICK dataset of
Marelli et al. (2014)from English into Dutch. Having a parallel inference
dataset allows us to compare both monolingual and multilingual NLP models for
English and Dutch on the two tasks. In the paper, we motivate and detail the
translation process, perform a baseline evaluation on both the original SICK
dataset and its Dutch incarnation SICK-NL, taking inspiration from Dutch
skipgram embeddings and contextualised embedding models. In addition, we
encapsulate two phenomena encountered in the translation to formulate stress
tests and verify how well the Dutch models capture syntactic restructurings
that do not affect semantics. Our main finding is all models perform worse on
SICK-NL than on SICK, indicating that the Dutch dataset is more challenging
than the English original. Results on the stress tests show that models don't
fully capture word order freedom in Dutch, warranting future systematic
studies.
| 2,021 |
Computation and Language
|
Persistent Anti-Muslim Bias in Large Language Models
|
It has been observed that large-scale language models capture undesirable
societal biases, e.g. relating to race and gender; yet religious bias has been
relatively unexplored. We demonstrate that GPT-3, a state-of-the-art contextual
language model, captures persistent Muslim-violence bias. We probe GPT-3 in
various ways, including prompt completion, analogical reasoning, and story
generation, to understand this anti-Muslim bias, demonstrating that it appears
consistently and creatively in different uses of the model and that it is
severe even compared to biases about other religious groups. For instance,
"Muslim" is analogized to "terrorist" in 23% of test cases, while "Jewish" is
mapped to "money" in 5% of test cases. We quantify the positive distraction
needed to overcome this bias with adversarial text prompts, and find that use
of the most positive 6 adjectives reduces violent completions for "Muslims"
from 66% to 20%, but which is still higher than for other religious groups.
| 2,021 |
Computation and Language
|
Persuasive Natural Language Generation -- A Literature Review
|
This literature review focuses on the use of Natural Language Generation
(NLG) to automatically detect and generate persuasive texts. Extending previous
research on automatic identification of persuasion in text, we concentrate on
generative aspects through conceptualizing determinants of persuasion in five
business-focused categories: benevolence, linguistic appropriacy, logical
argumentation, trustworthiness, tools and datasets. These allow NLG to increase
an existing message's persuasiveness. Previous research illustrates key aspects
in each of the above mentioned five categories. A research agenda to further
study persuasive NLG is developed. The review includes analysis of
seventy-seven articles, outlining the existing body of knowledge and showing
the steady progress in this research field.
| 2,021 |
Computation and Language
|
Interpretable Multi-Head Self-Attention model for Sarcasm Detection in
social media
|
Sarcasm is a linguistic expression often used to communicate the opposite of
what is said, usually something that is very unpleasant with an intention to
insult or ridicule. Inherent ambiguity in sarcastic expressions, make sarcasm
detection very difficult. In this work, we focus on detecting sarcasm in
textual conversations from various social networking platforms and online
media. To this end, we develop an interpretable deep learning model using
multi-head self-attention and gated recurrent units. Multi-head self-attention
module aids in identifying crucial sarcastic cue-words from the input, and the
recurrent units learn long-range dependencies between these cue-words to better
classify the input text. We show the effectiveness of our approach by achieving
state-of-the-art results on multiple datasets from social networking platforms
and online media. Models trained using our proposed approach are easily
interpretable and enable identifying sarcastic cues in the input text which
contribute to the final classification score. We visualize the learned
attention weights on few sample input texts to showcase the effectiveness and
interpretability of our model.
| 2,021 |
Computation and Language
|
KDLSQ-BERT: A Quantized Bert Combining Knowledge Distillation with
Learned Step Size Quantization
|
Recently, transformer-based language models such as BERT have shown
tremendous performance improvement for a range of natural language processing
tasks. However, these language models usually are computation expensive and
memory intensive during inference. As a result, it is difficult to deploy them
on resource-restricted devices. To improve the inference performance, as well
as reduce the model size while maintaining the model accuracy, we propose a
novel quantization method named KDLSQ-BERT that combines knowledge distillation
(KD) with learned step size quantization (LSQ) for language model quantization.
The main idea of our method is that the KD technique is leveraged to transfer
the knowledge from a "teacher" model to a "student" model when exploiting LSQ
to quantize that "student" model during the quantization training process.
Extensive experiment results on GLUE benchmark and SQuAD demonstrate that our
proposed KDLSQ-BERT not only performs effectively when doing different bit
(e.g. 2-bit $\sim$ 8-bit) quantization, but also outperforms the existing BERT
quantization methods, and even achieves comparable performance as the
full-precision base-line model while obtaining 14.9x compression ratio. Our
code will be public available.
| 2,021 |
Computation and Language
|
Hostility Detection and Covid-19 Fake News Detection in Social Media
|
Withtheadventofsocialmedia,therehasbeenanextremely rapid increase in the
content shared online. Consequently, the propagation of fake news and hostile
messages on social media platforms has also skyrocketed. In this paper, we
address the problem of detecting hostile and fake content in the Devanagari
(Hindi) script as a multi-class, multi-label problem. Using NLP techniques, we
build a model that makes use of an abusive language detector coupled with
features extracted via Hindi BERT and Hindi FastText models and metadata. Our
model achieves a 0.97 F1 score on coarse grain evaluation on Hostility
detection task. Additionally, we built models to identify fake news related to
Covid-19 in English tweets. We leverage entity information extracted from the
tweets along with textual representations learned from word embeddings and
achieve a 0.93 F1 score on the English fake news detection task.
| 2,021 |
Computation and Language
|
"Killing Me" Is Not a Spoiler: Spoiler Detection Model using Graph
Neural Networks with Dependency Relation-Aware Attention Mechanism
|
Several machine learning-based spoiler detection models have been proposed
recently to protect users from spoilers on review websites. Although dependency
relations between context words are important for detecting spoilers, current
attention-based spoiler detection models are insufficient for utilizing
dependency relations. To address this problem, we propose a new spoiler
detection model called SDGNN that is based on syntax-aware graph neural
networks. In the experiments on two real-world benchmark datasets, we show that
our SDGNN outperforms the existing spoiler detection models.
| 2,021 |
Computation and Language
|
Coarse-grained decomposition and fine-grained interaction for multi-hop
question answering
|
Recent advances regarding question answering and reading comprehension have
resulted in models that surpass human performance when the answer is contained
in a single, continuous passage of text, requiring only single-hop reasoning.
However, in actual scenarios, lots of complex queries require multi-hop
reasoning. The key to the Question Answering task is semantic feature
interaction between documents and questions, which is widely processed by
Bi-directional Attention Flow (Bi-DAF), but Bi-DAF generally captures only the
surface semantics of words in complex questions and fails to capture implied
semantic feature of intermediate answers. As a result, Bi-DAF partially ignores
part of the contexts related to the question and cannot extract the most
important parts of multiple documents. In this paper we propose a new model
architecture for multi-hop question answering, by applying two completion
strategies: (1) Coarse-Grain complex question Decomposition (CGDe) strategy are
introduced to decompose complex question into simple ones under the condition
of without any additional annotations (2) Fine-Grained Interaction (FGIn)
strategy are introduced to better represent each word in the document and
extract more comprehensive and accurate sentences related to the inference
path. The above two strategies are combined and tested on the SQuAD and
HotpotQA datasets, and the experimental results show that our method
outperforms state-of-the-art baselines.
| 2,021 |
Computation and Language
|
Walk in Wild: An Ensemble Approach for Hostility Detection in Hindi
Posts
|
As the reach of the internet increases, pejorative terms started flooding
over social media platforms. This leads to the necessity of identifying hostile
content on social media platforms. Identification of hostile contents on
low-resource languages like Hindi poses different challenges due to its diverse
syntactic structure compared to English. In this paper, we develop a simple
ensemble based model on pre-trained mBERT and popular classification algorithms
like Artificial Neural Network (ANN) and XGBoost for hostility detection in
Hindi posts. We formulated this problem as binary classification (hostile and
non-hostile class) and multi-label multi-class classification problem (for more
fine-grained hostile classes). We received third overall rank in the
competition and weighted F1-scores of ~0.969 and ~0.61 on the binary and
multi-label multi-class classification tasks respectively.
| 2,021 |
Computation and Language
|
Unstructured Knowledge Access in Task-oriented Dialog Modeling using
Language Inference, Knowledge Retrieval and Knowledge-Integrative Response
Generation
|
Dialog systems enriched with external knowledge can handle user queries that
are outside the scope of the supporting databases/APIs. In this paper, we
follow the baseline provided in DSTC9 Track 1 and propose three subsystems,
KDEAK, KnowleDgEFactor, and Ens-GPT, which form the pipeline for a
task-oriented dialog system capable of accessing unstructured knowledge.
Specifically, KDEAK performs knowledge-seeking turn detection by formulating
the problem as natural language inference using knowledge from dialogs,
databases and FAQs. KnowleDgEFactor accomplishes the knowledge selection task
by formulating a factorized knowledge/document retrieval problem with three
modules performing domain, entity and knowledge level analyses. Ens-GPT
generates a response by first processing multiple knowledge snippets, followed
by an ensemble algorithm that decides if the response should be solely derived
from a GPT2-XL model, or regenerated in combination with the top-ranking
knowledge snippet. Experimental results demonstrate that the proposed pipeline
system outperforms the baseline and generates high-quality responses, achieving
at least 58.77% improvement on BLEU-4 score.
| 2,021 |
Computation and Language
|
Hierarchical Multitask Learning with Dependency Parsing for Japanese
Semantic Role Labeling Improves Performance of Argument Identification
|
With the advent of FrameNet and PropBank, many semantic role labeling (SRL)
systems have been proposed in English. Although research on Japanese predicate
argument structure analysis (PASA) has been conducted, most studies focused on
surface cases. There are only few previous works on Japanese SRL for deep
cases, and their models' accuracies are low. Therefore, we propose a
hierarchical multitask learning method with dependency parsing (DP) and show
that our model achieves state-of-the-art results in Japanese SRL. Also, we
conduct experiments with a joint model that performs both argument
identification and argument classification simultaneously. The result suggests
that multitasking with DP is mainly effective for argument identification.
| 2,021 |
Computation and Language
|
The Impact of Post-editing and Machine Translation on Creativity and
Reading Experience
|
This article presents the results of a study involving the translation of a
fictional story from English into Catalan in three modalities:
machine-translated (MT), post-edited (MTPE) and translated without aid (HT).
Each translation was analysed to evaluate its creativity. Subsequently, a
cohort of 88 Catalan participants read the story in a randomly assigned
modality and completed a survey. The results show that HT presented a higher
creativity score if compared to MTPE and MT. HT also ranked higher in narrative
engagement, and translation reception, while MTPE ranked marginally higher in
enjoyment. HT and MTPE show no statistically significant differences in any
category, whereas MT does in all variables tested. We conclude that creativity
is highest when professional translators intervene in the process, especially
when working without any aid. We hypothesize that creativity in translation
could be the factor that enhances reading engagement and the reception of
translated literary texts.
| 2,021 |
Computation and Language
|
Empirical Evaluation of Supervision Signals for Style Transfer Models
|
Text style transfer has gained increasing attention from the research
community over the recent years. However, the proposed approaches vary in many
ways, which makes it hard to assess the individual contribution of the model
components. In style transfer, the most important component is the optimization
technique used to guide the learning in the absence of parallel training data.
In this work we empirically compare the dominant optimization paradigms which
provide supervision signals during training: backtranslation, adversarial
training and reinforcement learning. We find that backtranslation has
model-specific limitations, which inhibits training style transfer models.
Reinforcement learning shows the best performance gains, while adversarial
training, despite its popularity, does not offer an advantage over the latter
alternative. In this work we also experiment with Minimum Risk Training, a
popular technique in the machine translation community, which, to our
knowledge, has not been empirically evaluated in the task of style transfer. We
fill this research gap and empirically show its efficacy.
| 2,021 |
Computation and Language
|
TextGNN: Improving Text Encoder via Graph Neural Network in Sponsored
Search
|
Text encoders based on C-DSSM or transformers have demonstrated strong
performance in many Natural Language Processing (NLP) tasks. Low latency
variants of these models have also been developed in recent years in order to
apply them in the field of sponsored search which has strict computational
constraints. However these models are not the panacea to solve all the Natural
Language Understanding (NLU) challenges as the pure semantic information in the
data is not sufficient to fully identify the user intents. We propose the
TextGNN model that naturally extends the strong twin tower structured encoders
with the complementary graph information from user historical behaviors, which
serves as a natural guide to help us better understand the intents and hence
generate better language representations. The model inherits all the benefits
of twin tower models such as C-DSSM and TwinBERT so that it can still be used
in the low latency environment while achieving a significant performance gain
than the strong encoder-only counterpart baseline models in both offline
evaluations and online production system. In offline experiments, the model
achieves a 0.14% overall increase in ROC-AUC with a 1% increased accuracy for
long-tail low-frequency Ads, and in the online A/B testing, the model shows a
2.03% increase in Revenue Per Mille with a 2.32% decrease in Ad defect rate.
| 2,021 |
Computation and Language
|
Grid Search Hyperparameter Benchmarking of BERT, ALBERT, and LongFormer
on DuoRC
|
The purpose of this project is to evaluate three language models named BERT,
ALBERT, and LongFormer on the Question Answering dataset called DuoRC. The
language model task has two inputs, a question, and a context. The context is a
paragraph or an entire document while the output is the answer based on the
context. The goal is to perform grid search hyperparameter fine-tuning using
DuoRC. Pretrained weights of the models are taken from the Huggingface library.
Different sets of hyperparameters are used to fine-tune the models using two
versions of DuoRC which are the SelfRC and the ParaphraseRC. The results show
that the ALBERT (pretrained using the SQuAD1 dataset) has an F1 score of 76.4
and an accuracy score of 68.52 after fine-tuning on the SelfRC dataset. The
Longformer model (pretrained using the SQuAD and SelfRC datasets) has an F1
score of 52.58 and an accuracy score of 46.60 after fine-tuning on the
ParaphraseRC dataset. The current results outperformed the results from the
previous model by DuoRC.
| 2,021 |
Computation and Language
|
Weakly-Supervised Hierarchical Models for Predicting Persuasive
Strategies in Good-faith Textual Requests
|
Modeling persuasive language has the potential to better facilitate our
decision-making processes. Despite its importance, computational modeling of
persuasion is still in its infancy, largely due to the lack of benchmark
datasets that can provide quantitative labels of persuasive strategies to
expedite this line of research. To this end, we introduce a large-scale
multi-domain text corpus for modeling persuasive strategies in good-faith text
requests. Moreover, we design a hierarchical weakly-supervised latent variable
model that can leverage partially labeled data to predict such associated
persuasive strategies for each sentence, where the supervision comes from both
the overall document-level labels and very limited sentence-level labels.
Experimental results showed that our proposed method outperformed existing
semi-supervised baselines significantly. We have publicly released our code at
https://github.com/GT-SALT/Persuasion_Strategy_WVAE.
| 2,021 |
Computation and Language
|
Comparison of Machine Learning for Sentiment Analysis in Detecting
Anxiety Based on Social Media Data
|
All groups of people felt the impact of the COVID-19 pandemic. This situation
triggers anxiety, which is bad for everyone. The government's role is very
influential in solving these problems with its work program. It also has many
pros and cons that cause public anxiety. For that, it is necessary to detect
anxiety to improve government programs that can increase public expectations.
This study applies machine learning to detecting anxiety based on social media
comments regarding government programs to deal with this pandemic. This concept
will adopt a sentiment analysis in detecting anxiety based on positive and
negative comments from netizens. The machine learning methods implemented
include K-NN, Bernoulli, Decision Tree Classifier, Support Vector Classifier,
Random Forest, and XG-boost. The data sample used is the result of crawling
YouTube comments. The data used amounted to 4862 comments consisting of
negative and positive data with 3211 and 1651. Negative data identify anxiety,
while positive data identifies hope (not anxious). Machine learning is
processed based on feature extraction of count-vectorization and TF-IDF. The
results showed that the sentiment data amounted to 3889 and 973 in testing, and
training with the greatest accuracy was the random forest with feature
extraction of vectorization count and TF-IDF of 84.99% and 82.63%,
respectively. The best precision test is K-NN, while the best recall is
XG-Boost. Thus, Random Forest is the best accurate to detect someone's anxiety
based-on data from social media.
| 2,021 |
Computation and Language
|
Tuiteamos o pongamos un tuit? Investigating the Social Constraints of
Loanword Integration in Spanish Social Media
|
Speakers of non-English languages often adopt loanwords from English to
express new or unusual concepts. While these loanwords may be borrowed
unchanged, speakers may also integrate the words to fit the constraints of
their native language, e.g. creating Spanish "tuitear" from English "tweet."
Linguists have often considered the process of loanword integration to be more
dependent on language-internal constraints, but sociolinguistic constraints
such as speaker background remain only qualitatively understood. We investigate
the role of social context and speaker background in Spanish speakers' use of
integrated loanwords on social media. We find first that newspaper authors use
the integrated forms of loanwords and native words more often than social media
authors, showing that integration is associated with formal domains. In social
media, we find that speaker background and expectations of formality explain
loanword and native word integration, such that authors who use more Spanish
and who write to a wider audience tend to use integrated verb forms more often.
This study shows that loanword integration reflects not only language-internal
constraints but also social expectations that vary by conversation and speaker.
| 2,021 |
Computation and Language
|
To Understand Representation of Layer-aware Sequence Encoders as
Multi-order-graph
|
In this paper, we propose an explanation of representation for self-attention
network (SAN) based neural sequence encoders, which regards the information
captured by the model and the encoding of the model as graph structure and the
generation of these graph structures respectively. The proposed explanation
applies to existing works on SAN-based models and can explain the relationship
among the ability to capture the structural or linguistic information, depth of
model, and length of sentence, and can also be extended to other models such as
recurrent neural network based models. We also propose a revisited multigraph
called Multi-order-Graph (MoG) based on our explanation to model the graph
structures in the SAN-based model as subgraphs in MoG and convert the encoding
of SAN-based model to the generation of MoG. Based on our explanation, we
further introduce a Graph-Transformer by enhancing the ability to capture
multiple subgraphs of different orders and focusing on subgraphs of high
orders. Experimental results on multiple neural machine translation tasks show
that the Graph-Transformer can yield effective performance improvement.
| 2,023 |
Computation and Language
|
ComQA:Compositional Question Answering via Hierarchical Graph Neural
Networks
|
With the development of deep learning techniques and large scale datasets,
the question answering (QA) systems have been quickly improved, providing more
accurate and satisfying answers. However, current QA systems either focus on
the sentence-level answer, i.e., answer selection, or phrase-level answer,
i.e., machine reading comprehension. How to produce compositional answers has
not been throughout investigated. In compositional question answering, the
systems should assemble several supporting evidence from the document to
generate the final answer, which is more difficult than sentence-level or
phrase-level QA. In this paper, we present a large-scale compositional question
answering dataset containing more than 120k human-labeled questions. The answer
in this dataset is composed of discontiguous sentences in the corresponding
document. To tackle the ComQA problem, we proposed a hierarchical graph neural
networks, which represents the document from the low-level word to the
high-level sentence. We also devise a question selection and node selection
task for pre-training. Our proposed model achieves a significant improvement
over previous machine reading comprehension methods and pre-training methods.
Codes and dataset can be found at \url{https://github.com/benywon/ComQA}.
| 2,021 |
Computation and Language
|
Match-Ignition: Plugging PageRank into Transformer for Long-form Text
Matching
|
Neural text matching models have been widely used in community question
answering, information retrieval, and dialogue. However, these models designed
for short texts cannot well address the long-form text matching problem,
because there are many contexts in long-form texts can not be directly aligned
with each other, and it is difficult for existing models to capture the key
matching signals from such noisy data. Besides, these models are
computationally expensive for simply use all textual data indiscriminately. To
tackle the effectiveness and efficiency problem, we propose a novel
hierarchical noise filtering model, namely Match-Ignition. The main idea is to
plug the well-known PageRank algorithm into the Transformer, to identify and
filter both sentence and word level noisy information in the matching process.
Noisy sentences are usually easy to detect because previous work has shown that
their similarity can be explicitly evaluated by the word overlapping, so we
directly use PageRank to filter such information based on a sentence similarity
graph. Unlike sentences, words rely on their contexts to express concrete
meanings, so we propose to jointly learn the filtering and matching process, to
well capture the critical word-level matching signals. Specifically, a word
graph is first built based on the attention scores in each self-attention block
of Transformer, and key words are then selected by applying PageRank on this
graph. In this way, noisy words will be filtered out layer by layer in the
matching process. Experimental results show that Match-Ignition outperforms
both SOTA short text matching models and recent long-form text matching models.
We also conduct detailed analysis to show that Match-Ignition efficiently
captures important sentences and words, to facilitate the long-form text
matching process.
| 2,021 |
Computation and Language
|
Linguistically-Enriched and Context-Aware Zero-shot Slot Filling
|
Slot filling is identifying contiguous spans of words in an utterance that
correspond to certain parameters (i.e., slots) of a user request/query. Slot
filling is one of the most important challenges in modern task-oriented dialog
systems. Supervised learning approaches have proven effective at tackling this
challenge, but they need a significant amount of labeled training data in a
given domain. However, new domains (i.e., unseen in training) may emerge after
deployment. Thus, it is imperative that these models seamlessly adapt and fill
slots from both seen and unseen domains -- unseen domains contain unseen slot
types with no training data, and even seen slots in unseen domains are
typically presented in different contexts. This setting is commonly referred to
as zero-shot slot filling. Little work has focused on this setting, with
limited experimental evaluation. Existing models that mainly rely on
context-independent embedding-based similarity measures fail to detect slot
values in unseen domains or do so only partially. We propose a new zero-shot
slot filling neural model, LEONA, which works in three steps. Step one acquires
domain-oblivious, context-aware representations of the utterance word by
exploiting (a) linguistic features; (b) named entity recognition cues; (c)
contextual embeddings from pre-trained language models. Step two fine-tunes
these rich representations and produces slot-independent tags for each word.
Step three exploits generalizable context-aware utterance-slot similarity
features at the word level, uses slot-independent tags, and contextualizes them
to produce slot-specific predictions for each word. Our thorough evaluation on
four diverse public datasets demonstrates that our approach consistently
outperforms the SOTA models by 17.52%, 22.15%, 17.42%, and 17.95% on average
for unseen domains on SNIPS, ATIS, MultiWOZ, and SGD datasets, respectively.
| 2,021 |
Computation and Language
|
GENIE: Toward Reproducible and Standardized Human Evaluation for Text
Generation
|
While often assumed a gold standard, effective human evaluation of text
generation remains an important, open area for research. We revisit this
problem with a focus on producing consistent evaluations that are reproducible
-- over time and across different populations. We study this goal in different
stages of the human evaluation pipeline. In particular, we consider design
choices for the annotation interface used to elicit human judgments and their
impact on reproducibility. Furthermore, we develop an automated mechanism for
maintaining annotator quality via a probabilistic model that detects and
excludes noisy annotators. Putting these lessons together, we introduce GENIE:
a system for running standardized human evaluations across different generation
tasks. We instantiate GENIE with datasets representing four core challenges in
text generation: machine translation, summarization, commonsense reasoning, and
machine comprehension. For each task, GENIE offers a leaderboard that
automatically crowdsources annotations for submissions, evaluating them along
axes such as correctness, conciseness, and fluency. We have made the GENIE
leaderboards publicly available, and have already ranked 50 submissions from 10
different research groups. We hope GENIE encourages further progress toward
effective, standardized evaluations for text generation.
| 2,022 |
Computation and Language
|
Efficiently Fusing Pretrained Acoustic and Linguistic Encoders for
Low-resource Speech Recognition
|
End-to-end models have achieved impressive results on the task of automatic
speech recognition (ASR). For low-resource ASR tasks, however, labeled data can
hardly satisfy the demand of end-to-end models. Self-supervised acoustic
pre-training has already shown its amazing ASR performance, while the
transcription is still inadequate for language modeling in end-to-end models.
In this work, we fuse a pre-trained acoustic encoder (wav2vec2.0) and a
pre-trained linguistic encoder (BERT) into an end-to-end ASR model. The fused
model only needs to learn the transfer from speech to language during
fine-tuning on limited labeled data. The length of the two modalities is
matched by a monotonic attention mechanism without additional parameters.
Besides, a fully connected layer is introduced for the hidden mapping between
modalities. We further propose a scheduled fine-tuning strategy to preserve and
utilize the text context modeling ability of the pre-trained linguistic
encoder. Experiments show our effective utilizing of pre-trained modules. Our
model achieves better recognition performance on CALLHOME corpus (15 hours)
than other end-to-end models.
| 2,021 |
Computation and Language
|
Few Shot Dialogue State Tracking using Meta-learning
|
Dialogue State Tracking (DST) forms a core component of automated chatbot
based systems designed for specific goals like hotel, taxi reservation, tourist
information, etc. With the increasing need to deploy such systems in new
domains, solving the problem of zero/few-shot DST has become necessary. There
has been a rising trend for learning to transfer knowledge from resource-rich
domains to unknown domains with minimal need for additional data. In this work,
we explore the merits of meta-learning algorithms for this transfer and hence,
propose a meta-learner D-REPTILE specific to the DST problem. With extensive
experimentation, we provide clear evidence of benefits over conventional
approaches across different domains, methods, base models, and datasets with
significant (5-25%) improvement over the baseline in a low-data setting. Our
proposed meta-learner is agnostic of the underlying model and hence any
existing state-of-the-art DST system can improve its performance on unknown
domains using our training strategy.
| 2,021 |
Computation and Language
|
Narration Generation for Cartoon Videos
|
Research on text generation from multimodal inputs has largely focused on
static images, and less on video data. In this paper, we propose a new task,
narration generation, that is complementing videos with narration texts that
are to be interjected in several places. The narrations are part of the video
and contribute to the storyline unfolding in it. Moreover, they are
context-informed, since they include information appropriate for the timeframe
of video they cover, and also, do not need to include every detail shown in
input scenes, as a caption would. We collect a new dataset from the animated
television series Peppa Pig. Furthermore, we formalize the task of narration
generation as including two separate tasks, timing and content generation, and
present a set of models on the new task.
| 2,021 |
Computation and Language
|
What Makes Good In-Context Examples for GPT-$3$?
|
GPT-$3$ has attracted lots of attention due to its superior performance
across a wide range of NLP tasks, especially with its powerful and versatile
in-context few-shot learning ability. Despite its success, we found that the
empirical results of GPT-$3$ depend heavily on the choice of in-context
examples. In this work, we investigate whether there are more effective
strategies for judiciously selecting in-context examples (relative to random
sampling) that better leverage GPT-$3$'s few-shot capabilities. Inspired by the
recent success of leveraging a retrieval module to augment large-scale neural
network models, we propose to retrieve examples that are semantically-similar
to a test sample to formulate its corresponding prompt. Intuitively, the
in-context examples selected with such a strategy may serve as more informative
inputs to unleash GPT-$3$'s extensive knowledge. We evaluate the proposed
approach on several natural language understanding and generation benchmarks,
where the retrieval-based prompt selection approach consistently outperforms
the random baseline. Moreover, it is observed that the sentence encoders
fine-tuned on task-related datasets yield even more helpful retrieval results.
Notably, significant gains are observed on tasks such as table-to-text
generation (41.9% on the ToTTo dataset) and open-domain question answering
(45.5% on the NQ dataset). We hope our investigation could help understand the
behaviors of GPT-$3$ and large-scale pre-trained LMs in general and enhance
their few-shot capabilities.
| 2,021 |
Computation and Language
|
Joint Energy-based Model Training for Better Calibrated Natural Language
Understanding Models
|
In this work, we explore joint energy-based model (EBM) training during the
finetuning of pretrained text encoders (e.g., Roberta) for natural language
understanding (NLU) tasks. Our experiments show that EBM training can help the
model reach a better calibration that is competitive to strong baselines, with
little or no loss in accuracy. We discuss three variants of energy functions
(namely scalar, hidden, and sharp-hidden) that can be defined on top of a text
encoder, and compare them in experiments. Due to the discreteness of text data,
we adopt noise contrastive estimation (NCE) to train the energy-based model. To
make NCE training more effective, we train an auto-regressive noise model with
the masked language model (MLM) objective.
| 2,021 |
Computation and Language
|
Abstractive Opinion Tagging
|
In e-commerce, opinion tags refer to a ranked list of tags provided by the
e-commerce platform that reflect characteristics of reviews of an item. To
assist consumers to quickly grasp a large number of reviews about an item,
opinion tags are increasingly being applied by e-commerce platforms. Current
mechanisms for generating opinion tags rely on either manual labelling or
heuristic methods, which is time-consuming and ineffective. In this paper, we
propose the abstractive opinion tagging task, where systems have to
automatically generate a ranked list of opinion tags that are based on, but
need not occur in, a given set of user-generated reviews.
The abstractive opinion tagging task comes with three main challenges: (1)
the noisy nature of reviews; (2) the formal nature of opinion tags vs. the
colloquial language usage in reviews; and (3) the need to distinguish between
different items with very similar aspects. To address these challenges, we
propose an abstractive opinion tagging framework, named AOT-Net, to generate a
ranked list of opinion tags given a large number of reviews. First, a
sentence-level salience estimation component estimates each review's salience
score. Next, a review clustering and ranking component ranks reviews in two
steps: first, reviews are grouped into clusters and ranked by cluster size;
then, reviews within each cluster are ranked by their distance to the cluster
center. Finally, given the ranked reviews, a rank-aware opinion tagging
component incorporates an alignment feature and alignment loss to generate a
ranked list of opinion tags. To facilitate the study of this task, we create
and release a large-scale dataset, called eComTag, crawled from real-world
e-commerce websites. Extensive experiments conducted on the eComTag dataset
verify the effectiveness of the proposed AOT-Net in terms of various evaluation
metrics.
| 2,021 |
Computation and Language
|
Can a Fruit Fly Learn Word Embeddings?
|
The mushroom body of the fruit fly brain is one of the best studied systems
in neuroscience. At its core it consists of a population of Kenyon cells, which
receive inputs from multiple sensory modalities. These cells are inhibited by
the anterior paired lateral neuron, thus creating a sparse high dimensional
representation of the inputs. In this work we study a mathematical
formalization of this network motif and apply it to learning the correlational
structure between words and their context in a corpus of unstructured text, a
common natural language processing (NLP) task. We show that this network can
learn semantic representations of words and can generate both static and
context-dependent word embeddings. Unlike conventional methods (e.g., BERT,
GloVe) that use dense representations for word embedding, our algorithm encodes
semantic meaning of words and their context in the form of sparse binary hash
codes. The quality of the learned representations is evaluated on word
similarity analysis, word-sense disambiguation, and document classification. It
is shown that not only can the fruit fly network motif achieve performance
comparable to existing methods in NLP, but, additionally, it uses only a
fraction of the computational resources (shorter training time and smaller
memory footprint).
| 2,021 |
Computation and Language
|
Incremental Knowledge Based Question Answering
|
In the past years, Knowledge-Based Question Answering (KBQA), which aims to
answer natural language questions using facts in a knowledge base, has been
well developed. Existing approaches often assume a static knowledge base.
However, the knowledge is evolving over time in the real world. If we directly
apply a fine-tuning strategy on an evolving knowledge base, it will suffer from
a serious catastrophic forgetting problem. In this paper, we propose a new
incremental KBQA learning framework that can progressively expand learning
capacity as humans do. Specifically, it comprises a margin-distilled loss and a
collaborative exemplar selection method, to overcome the catastrophic
forgetting problem by taking advantage of knowledge distillation. We reorganize
the SimpleQuestion dataset to evaluate the proposed incremental learning
solution to KBQA. The comprehensive experiments demonstrate its effectiveness
and efficiency when working with the evolving knowledge base.
| 2,021 |
Computation and Language
|
HinFlair: pre-trained contextual string embeddings for pos tagging and
text classification in the Hindi language
|
Recent advancements in language models based on recurrent neural networks and
transformers architecture have achieved state-of-the-art results on a wide
range of natural language processing tasks such as pos tagging, named entity
recognition, and text classification. However, most of these language models
are pre-trained in high resource languages like English, German, Spanish.
Multi-lingual language models include Indian languages like Hindi, Telugu,
Bengali in their training corpus, but they often fail to represent the
linguistic features of these languages as they are not the primary language of
the study. We introduce HinFlair, which is a language representation model
(contextual string embeddings) pre-trained on a large monolingual Hindi corpus.
Experiments were conducted on 6 text classification datasets and a Hindi
dependency treebank to analyze the performance of these contextualized string
embeddings for the Hindi language. Results show that HinFlair outperforms
previous state-of-the-art publicly available pre-trained embeddings for
downstream tasks like text classification and pos tagging. Also, HinFlair when
combined with FastText embeddings outperforms many transformers-based language
models trained particularly for the Hindi language.
| 2,021 |
Computation and Language
|
Red Alarm for Pre-trained Models: Universal Vulnerability to
Neuron-Level Backdoor Attacks
|
Pre-trained models (PTMs) have been widely used in various downstream tasks.
The parameters of PTMs are distributed on the Internet and may suffer backdoor
attacks. In this work, we demonstrate the universal vulnerability of PTMs,
where fine-tuned PTMs can be easily controlled by backdoor attacks in arbitrary
downstream tasks. Specifically, attackers can add a simple pre-training task,
which restricts the output representations of trigger instances to pre-defined
vectors, namely neuron-level backdoor attack (NeuBA). If the backdoor
functionality is not eliminated during fine-tuning, the triggers can make the
fine-tuned model predict fixed labels by pre-defined vectors. In the
experiments of both natural language processing (NLP) and computer vision (CV),
we show that NeuBA absolutely controls the predictions for trigger instances
without any knowledge of downstream tasks. Finally, we apply several defense
methods to NeuBA and find that model pruning is a promising direction to resist
NeuBA by excluding backdoored neurons. Our findings sound a red alarm for the
wide use of PTMs. Our source code and models are available at
\url{https://github.com/thunlp/NeuBA}.
| 2,023 |
Computation and Language
|
Model Compression for Domain Adaptation through Causal Effect Estimation
|
Recent improvements in the predictive quality of natural language processing
systems are often dependent on a substantial increase in the number of model
parameters. This has led to various attempts of compressing such models, but
existing methods have not considered the differences in the predictive power of
various model components or in the generalizability of the compressed models.
To understand the connection between model compression and out-of-distribution
generalization, we define the task of compressing language representation
models such that they perform best in a domain adaptation setting. We choose to
address this problem from a causal perspective, attempting to estimate the
average treatment effect (ATE) of a model component, such as a single layer, on
the model's predictions. Our proposed ATE-guided Model Compression scheme
(AMoC), generates many model candidates, differing by the model components that
were removed. Then, we select the best candidate through a stepwise regression
model that utilizes the ATE to predict the expected performance on the target
domain. AMoC outperforms strong baselines on dozens of domain pairs across
three text classification and sequence tagging tasks.
| 2,021 |
Computation and Language
|
Neural Abstractive Text Summarizer for Telugu Language
|
Abstractive Text Summarization is the process of constructing semantically
relevant shorter sentences which captures the essence of the overall meaning of
the source text. It is actually difficult and very time consuming for humans to
summarize manually large documents of text. Much of work in abstractive text
summarization is being done in English and almost no significant work has been
reported in Telugu abstractive text summarization. So, we would like to propose
an abstractive text summarization approach for Telugu language using Deep
learning. In this paper we are proposing an abstractive text summarization Deep
learning model for Telugu language. The proposed architecture is based on
encoder-decoder sequential models with attention mechanism. We have applied
this model on manually created dataset to generate a one sentence summary of
the source text and have got good results measured qualitatively.
| 2,021 |
Computation and Language
|
Teach me how to Label: Labeling Functions from Natural Language with
Text-to-text Transformers
|
Annotated data has become the most important bottleneck in training accurate
machine learning models, especially for areas that require domain expertise. A
recent approach to deal with the above issue proposes using natural language
explanations instead of labeling individual data points, thereby increasing
human annotators' efficiency as well as decreasing costs substantially. This
paper focuses on the task of turning these natural language descriptions into
Python labeling functions by following a novel approach to semantic parsing
with pre-trained text-to-text Transformers. In a series of experiments our
approach achieves a new state of the art on the semantic parsing benchmark
CoNaLa, surpassing the previous best approach by 3.7 BLEU points. Furthermore,
on a manually constructed dataset of natural language descriptions-labeling
functions pairs we achieve a BLEU of 0.39. Our approach can be regarded as a
stepping stone towards models that are taught how to label in natural language,
instead of being provided specific labeled samples. Our code, constructed
dataset and models are available at
https://github.com/ypapanik/t5-for-code-generation.
| 2,021 |
Computation and Language
|
MONAH: Multi-Modal Narratives for Humans to analyze conversations
|
In conversational analyses, humans manually weave multimodal information into
the transcripts, which is significantly time-consuming. We introduce a system
that automatically expands the verbatim transcripts of video-recorded
conversations using multimodal data streams. This system uses a set of
preprocessing rules to weave multimodal annotations into the verbatim
transcripts and promote interpretability. Our feature engineering contributions
are two-fold: firstly, we identify the range of multimodal features relevant to
detect rapport-building; secondly, we expand the range of multimodal
annotations and show that the expansion leads to statistically significant
improvements in detecting rapport-building.
| 2,021 |
Computation and Language
|
Automatic punctuation restoration with BERT models
|
We present an approach for automatic punctuation restoration with BERT models
for English and Hungarian. For English, we conduct our experiments on Ted
Talks, a commonly used benchmark for punctuation restoration, while for
Hungarian we evaluate our models on the Szeged Treebank dataset. Our best
models achieve a macro-averaged $F_1$-score of 79.8 in English and 82.2 in
Hungarian. Our code is publicly available.
| 2,021 |
Computation and Language
|
Grounding Language to Entities and Dynamics for Generalization in
Reinforcement Learning
|
We investigate the use of natural language to drive the generalization of
control policies and introduce the new multi-task environment Messenger with
free-form text manuals describing the environment dynamics. Unlike previous
work, Messenger does not assume prior knowledge connecting text and state
observations $-$ the control policy must simultaneously ground the game manual
to entity symbols and dynamics in the environment. We develop a new model, EMMA
(Entity Mapper with Multi-modal Attention) which uses an entity-conditioned
attention module that allows for selective focus over relevant descriptions in
the manual for each entity in the environment. EMMA is end-to-end
differentiable and learns a latent grounding of entities and dynamics from text
to observations using only environment rewards. EMMA achieves successful
zero-shot generalization to unseen games with new dynamics, obtaining a 40%
higher win rate compared to multiple baselines. However, win rate on the
hardest stage of Messenger remains low (10%), demonstrating the need for
additional work in this direction.
| 2,021 |
Computation and Language
|
Exploring Lexical Irregularities in Hypothesis-Only Models of Natural
Language Inference
|
Natural Language Inference (NLI) or Recognizing Textual Entailment (RTE) is
the task of predicting the entailment relation between a pair of sentences
(premise and hypothesis). This task has been described as a valuable testing
ground for the development of semantic representations, and is a key component
in natural language understanding evaluation benchmarks. Models that understand
entailment should encode both, the premise and the hypothesis. However,
experiments by Poliak et al. revealed a strong preference of these models
towards patterns observed only in the hypothesis, based on a 10 dataset
comparison. Their results indicated the existence of statistical irregularities
present in the hypothesis that bias the model into performing competitively
with the state of the art. While recast datasets provide large scale generation
of NLI instances due to minimal human intervention, the papers that generate
them do not provide fine-grained analysis of the potential statistical patterns
that can bias NLI models. In this work, we analyze hypothesis-only models
trained on one of the recast datasets provided in Poliak et al. for word-level
patterns. Our results indicate the existence of potential lexical biases that
could contribute to inflating the model performance.
| 2,021 |
Computation and Language
|
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.