Improved ASR for English-isiZulu code-switched speech with semi-supervised training.
problem Improving ASR for code-switched speech between English and isiZulu.
method Semi-supervised training using automatic transcription of multilingual speech data.
result Semi-supervised training achieved significant WER reduction in ASR performance.
Study examines bias in language models across multiple languages.
problem Assessing bias in language models across different languages.
method Semi-automatically translated data sets into multiple languages, analyzed mono- and multilingual models.
result Notable differences in bias across languages, with Turkish models showing least stereotypes.
In this paper, we attempt to improve Statistical Machine Translation (SMT) systems on a very diverse set of language pairs (in both directions): Czech - English, Vietnamese - English, French - English and German - English. To accomplish this, we performed translation model training, created adaptations of training sett…
Proposes a multilingual email segmentation benchmark and model.
problem Lack of multilingual email zoning corpora and models.
method Analysis of existing corpora, development of multilingual benchmark, introduction of OKAPI model.
result OKAPI model achieves state-of-the-art performance in English and generalizes well to unseen languages.
This research explores the effects of various training settings on a Polish to English Statistical Machine Translation system for spoken language. Various elements of the TED, Europarl, and OPUS parallel text corpora were used as the basis for training of language models, for development, tuning and testing of the tran…
Automates translating natural language to Verilog for digital design.
problem Manual translation of natural language specifications to Verilog is time-consuming and error-prone.
method Fine-tuned GPT-2 to derive Verilog from English, using a dataset of design tasks.
result GPT-2 achieved 94.8% correct translation across simple and abstract design tasks.
RNN models classify poem meters from plain text with high accuracy.
problem Classifying poem meters from plain text.
method Character-level encoding, RNN models, no feature handcrafting.
result 96.38% accuracy for Arabic poems, 82.31% for English poems.
Improved neural language models using adversarial training.
problem Overfitting in large-scale neural language models.
method Adversarial training mechanism to introduce adversarial noise to output embedding layers.
result Improved perplexity scores and BLEU scores on various language tasks.
We describe a unified and coherent syntactic framework for supporting a semantically-informed syntactic approach to statistical machine translation. Semantically enriched syntactic tags assigned to the target-language training texts improved translation quality. The resulting system significantly outperformed a linguis…
This research explores the effects of various training settings from Polish to English Statistical Machine Translation system for spoken language. Various elements of the TED parallel text corpora for the IWSLT 2013 evaluation campaign were used as the basis for training of language models, and for development, tuning …
BERT model improves cross-lingual document retrieval.
problem Improving cross-lingual information retrieval between English queries and foreign documents.
method Introduced a deep relevance matching model based on BERT, trained with weak supervision using CLIR data.
result Model outperforms competitive baseline approaches in retrieving Lithuanian documents against short English queries.
This paper tackles rare word problem in low-resource language pairs using NMT.
problem Rare word problem in neural machine translation, especially for low-resource languages.
method Three solutions: enhanced source context, morphology learning, and wordnet synonyms.
result Significant improvements in BLEU scores (+1.0 points) on English-Vietnamese and Japanese-Vietnamese.
Develops a cross-lingual hate speech detection model using pre-trained Transformers.
problem Detecting hate speech in low-resource languages.
method Utilizes frozen Transformer language models and AXEL attention-based classification block for zero-shot and few-shot learning.
result Demonstrates highly competitive results on English and Spanish subsets of the HatEval challenge.
MGLM models all possible language channel factorizations for improved multilingual generation.
problem Generating multilingual text with flexibility and quality.
method Generative joint distribution model over language channels, marginalizing all possible factorizations.
result MGLM outperforms traditional models in multilingual generation tasks.
We show that the predictability of letters in written English texts depends strongly on their position in the word. The first letters are usually the least easy to predict. This agrees with the intuitive notion that words are well defined subunits in written languages, with much weaker correlations across these units t…
Dataset for measuring reading levels in India's children.
problem Measuring reading levels in India's vast population.
method Developed ASER dataset with 5,301 subjects in Hindi, Marathi, and English.
result Achieved 86% accuracy in English language classification.
In this paper, we present Neural Phrase-based Machine Translation (NPMT). Our method explicitly models the phrase structures in output sequences using Sleep-WAke Networks (SWAN), a recently proposed segmentation-based sequence modeling method. To mitigate the monotonic alignment requirement of SWAN, we introduce a new …
Investigates neural TTS systems for Japanese and English.
problem Improving neural TTS systems for high-quality speech synthesis.
method Comparative study of neural sequence-to-sequence TTS vs. DNN pipeline TTS, varying model architecture, parameter size, and language.
result A neural sequence-to-sequence TTS system requires sufficient model parameters and a powerful encoder for high-quality speech synthesis.
Recent works have highlighted the strength of the Transformer architecture on sequence tasks while, at the same time, neural architecture search (NAS) has begun to outperform human-designed models. Our goal is to apply NAS to search for a better alternative to the Transformer. We first construct a large search space in…
Improved neural model predicts gender from tweets.
problem Predicting gender from Twitter text.
method RNN model with attention, LSA-reduced n-gram features.
result Improved model achieves state-of-the-art performance on English tweets.
Universal Language Model for Fine-tuning [arXiv:1801.06146] (ULMFiT) is one of the first NLP methods for efficient inductive transfer learning. Unsupervised pretraining results in improvements on many NLP tasks for English. In this paper, we describe a new method that uses subword tokenization to adapt ULMFiT to langua…
Given that South African education is in crisis, strategies for improvement and sustainability of high-quality, up-to-date education must be explored. In the migration of education online, inclusion of machine translation for low-resourced local languages becomes necessary. This paper aims to spur the use of current ne…
Word2Vec embedding for Italian language developed.
problem Lack of word embedding for Italian language.
method Parameter tuning of Word2Vec for Italian language.
result An embedding for Italian language created.
A deep learning approach generates math word problems in multiple languages.
problem Template-based mechanisms for generating mathematical word problems lack customizability and creativity.
method Character Level Long Short Term Memory Network (LSTM) and POS tags are used to generate and resolve constraints in generated problems.
result The approach generates accurate math word problems in English and Sinhala with over 90% accuracy.
The task of determining a speaker's native language based only on his speeches in a second language is known as Native Language Identification or NLI. Due to its increasing applications in various domains of speech signal processing, this has emerged as an important research area in recent times. In this paper we have …
Improved sentiment analysis in Korean finance using masked PLMs.
problem Fine-grained sentiment analysis in non-English finance literature is lacking due to limited annotated data.
method Developed KorFinASC dataset and applied TGT-Masking to PLMs to remove non-stationary knowledge.
result Improved classification accuracy by 22.63% on KorFinASC.
Study finds common poetic themes across languages over time.
problem Understanding thematic evolution in different poetic traditions.
method Applied Latent Dirichlet Allocation (LDA) to poetry corpora of four languages.
result Identified common themes and their temporal trends across poetic traditions.
System identifies language of transliterated text.
problem Users struggle to understand non-native language transliterated text.
method Feature extraction of phonetic syllables using LSTM network.
result System accurately identifies language of transliterated text.
Proposes using Wasserstein barycenter for better multilingual alignment.
problem Finding word-to-word translations between multiple languages without parallel data.
method Uses Wasserstein barycenter as a more informative pivot language, minimizing pairwise transportation costs.
result Demonstrates state-of-the-art performances on standard benchmarks.
We describe a simple neural language model that relies only on character-level inputs. Predictions are still made at the word-level. Our model employs a convolutional neural network (CNN) and a highway network over characters, whose output is given to a long short-term memory (LSTM) recurrent neural network language mo…
Neural machine translation models trained for 5 South African languages.
problem Lack of resources and research for machine translation in African languages.
method Training neural machine translation models for 5 South African languages using modern techniques.
result Promises of neural machine translation for African languages.
This new research explores the effects of various training methods on a Polish to English Statistical Machine Translation system for medical texts. Various elements of the EMEA parallel text corpora from the OPUS project were used as the basis for training of phrase tables and language models and for development, tunin…
End-to-end Sanskrit TTS developed with limited data, achieving good quality.
problem Developing natural-sounding speech for Sanskrit with scarce data.
method Fine-tuning Tacotron2 model with WaveGlow and transfer learning.
result Achieved an overall MOS of 3.38 from 37 evaluators.
Real-time spell checker adapts to new languages.
problem No real-time, language-adaptable spell checkers for non-English languages.
method Used Wikipedia and subtitles data to generate dictionaries, created noisy channel datasets, compared with industry tools.
result System performs well across 24 languages, outperforming existing tools.
The dominant language models (LMs) such as n-gram and neural network (NN) models represent sentence probabilities in terms of conditionals. In contrast, a new trans-dimensional random field (TRF) LM has been recently introduced to show superior performances, where the whole sentence is modeled as a random field. In thi…
Pars-ABSA dataset for Persian aspect-based sentiment analysis.
problem Lack of public dataset for Persian aspect-based sentiment analysis.
method Manually annotated dataset with 5,114 positive, 3,061 negative, and 1,827 neutral samples.
result State-of-the-art performance of deep learning methods on Pars-ABSA compared to similar English datasets.
OT domain adaptation improves aphasia detection across languages.
problem Detecting aphasia in low-resource languages with limited data.
method Utilized OT domain adaptation to map linguistic features across multiple languages.
result OT domain adaptation significantly improved F1 scores for French and Mandarin aphasia detection.
Unsupervised RNNGs perform similarly to supervised ones in language modeling and grammar induction.
problem Training RNNGs requires annotated data, which is costly.
method Amortized variational inference with a neural CRF parser.
result Unsupervised RNNGs achieve comparable performance to supervised ones.
Large language models predict human sensory judgments across multiple modalities.
problem Determining the extent of perceptual information in language.
method State-of-the-art large language models were used to predict sensory judgments across six psychophysical datasets.
result Large language models can predict human sensory judgments across multiple modalities with significant correlation to human data.
In this paper, we present Russian language datasets in the digital humanities domain for the evaluation of word embedding techniques or similar language modeling and feature learning algorithms. The datasets are split into two task types, word intrusion and word analogy, and contain 31362 task units in total. The chara…
An NMT system for Indic languages outperforms Google Translate.
problem Challenges in translating Indic languages efficiently.
method Encoder-decoder with attention mechanism for neural machine translation.
result Outperforms Google Translate with a 6 BLEU score margin on English-Gujarati translation.
Transformer models show distinct spectral fingerprints under voice changes.
problem Detecting architectural biases in transformer models.
method Spectral analysis of attention-induced token graphs.
result Clear architectural signatures in model fingerprints correlate with language-specific behavior.
This paper reviews sentiment analysis on Indian languages.
problem Understanding sentiment in multilingual web data.
method Reviews and discusses approaches for sentiment analysis on Indian languages.
result Challenges in sentiment analysis on indigenous languages.
Project classifies Hinglish social content on platforms like Twitter, Reddit.
problem Classifying abusive and hate-inducing content in Hinglish on social media.
method Used deep learning with bi-directional sequence models and text augmentation techniques.
result Produced a state-of-the-art classifier that outperforms previous work.
Wi-Fringe recognizes named gestures using WiFi CSI without training data.
problem Lack of training data for WiFi-based gesture recognition.
method WiFi CSI values paired with activity names for recognition.
result Wi-Fringe detects activities with or without training data.
New models improve morpheme segmentation in low-resource languages.
problem Improving morpheme segmentation in low-resource languages.
method Two new models: LSTM pointer-generator and sequence-to-sequence with hard monotonic attention.
result Novel models outperform existing ones by up to 11.4% accuracy in low-resource settings.
New method evaluates LLMs fairness in universal prediction.
problem Evaluating fairness of large language models in universal prediction.
method Introducing batch regret as a modification of average regret for LLMs.
result Asymptotical value of batch regret for add-constant predictors on memoryless and first-order Markov sources.
Many methods have been used to recognize author personality traits from text, typically combining linguistic feature engineering with shallow learning models, e.g. linear regression or Support Vector Machines. This work uses deep-learning-based models and atomic features of text, the characters, to build hierarchical, …