Model learns multi-sense word embeddings using bilingual and monolingual data.
problem Learning multi-sense word embeddings with crosslingual information.
method Discrete autoencoder with encoder and decoder components.
result Bilingual word representations outperform monolingual ones across tasks.
Discriminative model improves bilingual lexicon induction.
problem Inducing accurate bilingual lexicons between languages.
method Combines bipartite matching and representation-based approaches with an efficient Viterbi EM algorithm.
result The prior improves the induced bilingual lexicons.
Geometric approach learns bilingual mappings from monolingual embeddings.
problem Bilingual lexicon induction and cross-lingual word similarity.
method Decouples learning into rotations and a metric, modeled as optimization on Riemannian manifolds.
result Outperforms previous approaches on bilingual lexicon induction and cross-lingual word similarity tasks.
We introduce BilBOWA (Bilingual Bag-of-Words without Alignments), a simple and computationally-efficient model for learning bilingual distributed representations of words which can scale to large monolingual datasets and does not require word-aligned parallel training data. Instead it trains directly on monolingual dat…
Unsupervised MT struggles with morphologically rich languages.
problem Limitations of unsupervised machine translation on morphologically rich languages.
method Adversarial unsupervised alignment of word embedding spaces for bilingual dictionary induction.
result A simple trick exploiting weak supervision from identical words improves unsupervised bilingual dictionary induction performance.
Our research extends the Bilingual Evaluation Understudy (BLEU) evaluation technique for statistical machine translation to make it more adjustable and robust. We intend to adapt it to resemble human evaluation more. We perform experiments to evaluate the performance of our technique against the primary existing evalua…
Simple framework decouples word alignment and multilingual embedding mapping.
problem Learning multilingual embeddings without supervision.
method Two-stage approach: 1) unsupervised word alignment, 2) mapping embeddings to shared space.
result Robust performance across various multilingual tasks, including distant languages.
Paper refines cross-lingual word embeddings using Manhattan norm.
problem Sensitivity of ℓ2 norm loss function to outliers in CLWEs. method Post-processing step using ℓ1 norm to improve CLWEs. result The ℓ1 refinement substantially outperforms state-of-the-art baselines. Cross-language learning allows us to use training data from one language to build models for a different language. Many approaches to bilingual learning require that we have word-level alignment of sentences from parallel corpora. In this work we explore the use of autoencoder-based methods for cross-language learning …
This paper discusses the effect of hubness in zero-shot learning, when ridge regression is used to find a mapping between the example space to the label space. Contrary to the existing approach, which attempts to find a mapping from the example space to the label space, we show that mapping labels into the example spac…
This research tackles automatic paraphrasing without translation.
problem Automatic paraphrasing without translation.
method Proposes a residual variant of vector-quantized variational auto-encoder trained on an unlabeled monolingual corpus.
result Monolingual paraphrasing outperforms unsupervised translation in all settings.
Geometric approach for unsupervised word embedding alignment.
problem Learning alignment between word embeddings of source and target languages.
method Formulates alignment as domain adaptation on the manifold of doubly stochastic matrices, employing Riemannian conjugate gradient algorithm.
result Empirically outperforms state-of-the-art methods on bilingual lexicon induction tasks.
Parallel sentences are a relatively scarce but extremely useful resource for many applications including cross-lingual retrieval and statistical machine translation. This research explores our new methodologies for mining such data from previously obtained comparable corpora. The task is highly practical since non-para…
When digitizing a print bilingual dictionary, whether via optical character recognition or manual entry, it is inevitable that errors are introduced into the electronic version that is created. We investigate automating the process of detecting errors in an XML representation of a digitized print dictionary using a hyb…
DeepSubQE estimates translation quality for subtitles, improving on existing methods.
problem Hard quality estimation for subtitle translations due to language variations.
method Proposes DeepSubQE, a hybrid network combining semantic and syntactic features.
result DeepSubQE outperforms existing methods by significant margin.
Improved unsupervised word translation using adversarial autoencoder with cycle consistency and input reconstruction.
problem Challenging language pairs and lack of parallel data for unsupervised word translation.
method Adversarial autoencoder with cycle consistency and input reconstruction regularization.
result More stable and better performance than recent approaches.
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.
This work decouples language from math problems to enable cross-language learning.
problem Current machine learning representations are language dependent.
method Inspired by linguistics, the work learns language agnostic representations.
result Models trained on one language achieve similar accuracies in other languages.
The multilingual nature of the world makes translation a crucial requirement today. Parallel dictionaries constructed by humans are a widely-available resource, but they are limited and do not provide enough coverage for good quality translation purposes, due to out-of-vocabulary words and neologisms. This motivates th…
Transformers trained with three normalization changes outperform state-of-the-art on low-resource translation tasks.
problem Improving the training of Transformers, especially on low-resource datasets.
method Three normalization changes: PreNorm, ScaleNorm, and FixNorm.
result Significant improvements in BLEU scores on low-resource translation tasks.
REALFIN benchmarks financial reasoning by removing implicit assumptions, revealing model weaknesses.
problem Models struggle when implicit assumptions are missing, leading to incorrect answers.
method Developed a bilingual benchmark that systematically removes essential premises from financial questions.
result General-purpose models over-commit, while finance-specialized models fail to identify missing premises.
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.
New method aligns word embeddings without supervision.
problem Aligning word embeddings trained on monolingual data.
method Wasserstein Procrustes approach with orthogonal and permutation matrices.
result Obtains state-of-the-art results in unsupervised word translation.
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.
In this work, a novel solution to the speaker identification problem is proposed through minimization of statistical divergences between the probability distribution (g). of feature vectors from the test utterance and the probability distributions of the feature vector corresponding to the speaker classes. This approac…
Paper develops multilingual job classification for ISCO and KZiS.
problem Classifying job advertisements for accurate occupation coding.
method Transformer architecture for hierarchical multi-class classification.
result Hierarchical structure improves prediction accuracy by 1-2 percentage points.
NMIXX fine-tunes embeddings for finance, outperforming general models in Korean.
problem Financial embeddings struggle in low-resource languages like Korean.
method Fine-tuned with 18.8K triplets, hard negatives, and translations.
result NMIXX achieves gains of +0.10 on English FinSTS and +0.22 on KorFinSTS.
Framework improves answer quality for diverse question phrasings.
problem Different phrasings of similar questions lead to varying answer quality.
method Constrained Mixed-Strategy GroupDRO for optimizing system-prompt weights.
result Reduces worst-case quality loss by 13.7% across metrics and groups.
AI generates a sequence of death causes from hospital records.
problem Accurate death reporting for vital statistics and policy formulation.
method Neural machine translation models to generate causal chains, incorporating medical domain knowledge.
result Achieved 16.04 BLEU score for generating accurate causal chains.