Is this you? Claim this profile to correct it, add a bio and choose the work people see first.
Claim this profileWorks125 from public data
- Modeling Coverage for Neural Machine Translation786
This paper proposes coverage-based NMT, which maintains a coverage vector to keep track of the attention history and improves both translation quality and alignment quality over standard attention- based NMT.
- Tree-to-string alignment template for statistical machine translation394
A novel translation model based on tree-to-string alignment template (TAT) which describes the alignment between a source parse tree and a target string that significantly outperforms Pharaoh, a state-of-the-art decoder for phrase-based models.
- Improving the Transformer Translation Model with Document-Level Context274
This work extends the Transformer model with a new context encoder to represent document-level context, which is then incorporated into the original encoder and decoder, and introduces a two-step training method to take full advantage of abundant sentence-level parallel corpora and limited document- level parallel Corpora.
- Neural Machine Translation with Reconstruction212
Experiments show that the proposed framework significantly improves the adequacy of NMT output and achieves superior translation result over state-of-the-art NMT and statistical MT systems.
- 207
- Learning to Remember Translation History with a Continuous Cache189
This work proposes to augment NMT models with a very light-weight cache-like memory network, which stores recent hidden representations as translation history and the probability distribution over generated words is updated online depending on the translation history retrieved from the memory.
- Visualizing and Understanding Neural Machine Translation186
This work proposes to use layer-wise relevance propagation (LRP) to compute the contribution of each contextual word to arbitrary hidden states in the attention-based encoder-decoder framework and shows that visualization with LRP helps to interpret the internal workings of NMT and analyze translation errors.
- Towards Robust Neural Machine Translation169
Experimental results on Chinese-English, English-German and English-French translation tasks show that the proposed approaches can not only achieve significant improvements over strong NMT systems but also improve the robustness of NMT models.
- Learning Structured Text Representations156
A model that can encode a document while automatically inducing rich structural dependencies is proposed that embeds a differentiable non-projective parsing algorithm into a neural model and uses attention mechanisms to incorporate the structural biases.
- Neural machine translation: A review of methods, resources, and tools152
A broad review of the methods for NMT is provided and focus on methods relating to architectures, decoding, and data augmentation, with a discussion of possible future research directions.
- Context Gates for Neural Machine Translation152
Context gates are proposed which dynamically control the ratios at which source and target contexts contribute to the generation of target words and can enhance both the adequacy and fluency of NMT with more careful control of the information flow from contexts.
- Skywork: A More Open Bilingual Foundation Model135
This bilingual foundation model is the most extensively trained and openly published LLMs of comparable size to date, and introduces a two-stage training methodology using a segmented corpus, targeting general purpose training and then domain-specific enhancement training, respectively.
- 110
- THUMT: An Open-Source Toolkit for Neural Machine Translation95
Experiments on English-German and Chinese-English datasets show that THUMT can obtain results that are comparable to state-of-the-art NMT systems.
- On the Inference Calibration of Neural Machine Translation94
This work provides in-depth analyses of the correlation between calibration and translation performance as well as linguistic properties of miscalibration and reports a number of interesting findings that might help humans better analyze, understand and improve NMT models.
- Reducing Word Omission Errors in Neural Machine Translation: A Contrastive Learning Approach87
This work proposes a contrastive learning approach to reducing word omission errors in NMT that achieves better translation performance than three baseline methods and designs different types of negative examples depending on the number of omitted words, word frequency, and part of speech.
- Improving Back-Translation with Uncertainty-based Confidence Estimation86
This work proposes to quantify the confidence of NMT model predictions based on model uncertainty, and shows that uncertainty-based confidence estimation significantly improves the performance of back-translation.
- Analysis of the Paragraph Vector Model for Information Retrieval84
The unregulated training process of PV is vulnerable to short document over-fitting that produces length bias in the final retrieval model, and the corpus-based negative sampling of PV leads to a weighting scheme for words that overly suppresses the importance of frequent words.
- Multi-Domain Neural Machine Translation with Word-Level Domain Context Discrimination81
This paper jointly model NMT with monolingual attention-based domain classification tasks and improves NMT as follows, distinguishing and exploiting word-level domain contexts for multi-domain NMT.
- Automated Essay Scoring with Discourse-Aware Neural Models79
Experiments on three essay scoring tasks show benefits from all three strategies in different combinations, with simpler architectures being more effective when less training data is available.
- 79
- 75
- Prior Knowledge Integration for Neural Machine Translation using Posterior Regularization71
This work proposes to use posterior regularization to provide a general framework for integrating prior knowledge into neural machine translation, and represents prior knowledge sources as features in a log-linear model, which guides the learning processing of the neural translation model.
- Recursive Autoencoders for ITG-Based Translation68
This work proposes to use recursive autoencoders to make full use of the entire merging blocks alternatively to enable predicting orders to exploit syntactic and semantic information from a neural language modeling’s perspective.
- Accurate Word Alignment Induction from Neural Machine Translation66
It is shown that attention weights do capture accurate word alignment, which could only be revealed if the correct decoding step and layer to induce word alignment is chosen, and two simple but effective interpretation methods for word alignment induction are presented.
- 61
- MSP: Multi-Stage Prompting for Making Pre-trained Language Models Better Translators60
This work presents Multi-Stage Prompting, a simple and automatic approach for leveraging pre-trained language models to translation tasks that can significantly improve the translation performance of pre- trained language models.
- Exploring Fine-grained Entity Type Constraints for Distantly Supervised Relation Extraction53
A novel method to explore fine-grained entity type constraints is proposed, and a series of methods to integrate the constraints with the relation extracting model are studied to achieve better precision/recall curves in sentential extraction with smoother curves in aggregated extraction which mean more stable models.
- Learning to Relate to Previous Turns in Conversational Search51
This paper uses a pseudo-labeling approach to annotate useful historical queries based on their impact on the retrieval results and proposes a multi-task learning framework to jointly train the selector and the retriever during fine-tuning.
- 50
- Zero-Resource Neural Machine Translation with Multi-Agent Communication Game49
This work proposes an interactive multimodal framework for zero-resource neural machine translation, where learners engage in cooperative image description games, and thus develop their own image captioning or neural machinetranslation model from the need to communicate in order to succeed at the game.
- Automatic prosody prediction for Chinese speech synthesis using BLSTM-RNN and embedding features49
This paper proposes to use neural networks to predict prosodic boundary labels directly from Chinese characters without any feature engineering, and shows that stacking feed-forward and bidirectional long short-term memory (BLSTM) recurrent network layers achieves superior performance over the CRF-based method.
- Discriminative Word Alignment by Linear Modeling48
A discriminative framework for word alignment based on a linear model that achieves state-of-the-art alignment quality on three word alignment shared tasks for five language pairs with varying divergence and richness of resources and improves translation performance for various statistical machine translation systems.
- 47
- 46
- 45
- Peptide design through binding interface mimicry with PepMimic39
An artificial intelligence algorithm, PepMimic, is developed to transform a known receptor or an existing antibody of a target into a short peptide binder by mimicking the binding interfaces between targets and known binders.
- Building Earth Mover's Distance on Bilingual Word Embeddings for Machine Translation38
Earth Mover's Distance is introduced by providing a natural formulation that translates words in a holistic fashion, addressing the limitations of the nearest neighbor, and extended to a new task of identifying parallel sentences, useful for statistical machine translation systems, thereby expanding the application realm of bilingual word embeddings.
- A Neural Reordering Model for Phrase-based Translation38
A neural reordering model that conditions reordering probabilities on the words of both the current and previous phrase pairs significantly improves context sensitivity and reduces reordering ambiguity.
- Mask-Align: Self-Supervised Neural Word Alignment37
The proposed Mask-Align is a self-supervised word alignment model that takes advantage of the full context on the target side, and introduces an attention variant called leaky attention, which alleviates the problem of unexpected high cross-attention weights on special tokens such as periods.
- Exploiting reverse target-side contexts for neural machine translation via asynchronous bidirectional decoding37
This paper extends the conventional attentional encoder-decoder NMT framework by introducing a backward decoder, in order to explore asynchronous bidirectional decoding for NMT, and achieves significant improvements over the conventional RNNSearch.
- 36
- 34
- Neural Machine Translation With Explicit Phrase Alignment34
The key idea is to build a search space similar to that of phrase-based statistical machine translation for neural machine translation where phrase alignment is readily available and a new decoding algorithm is designed that can easily impose lexical and structural constraints.
- Incorporating Sememes into Chinese Definition Modeling34
AAM successfully incorporates sememes for generating the definition with an adaptive attention mechanism and SAAM further replaces recurrent connections in AAM with self-attention and relies entirely on the attention mechanism, reducing the path length between word, sememe and definition.
- End-to-End Segmentation-based News Summarization30
A novel segmentation-based language generation model adapted from pre-trained language models that can jointly segment a document and produce the summary for each section is proposed.
- Maximum entropy based rule selection model for syntax-based statistical machine translation30
This paper incorporates the MERS model into a state-of-the-art linguistically syntax-based SMT model, the tree-to-string alignment template model, and shows that the approach achieves significant improvements over the baseline system.
- DiPair: Fast and Accurate Distillation for Trillion-Scale Text Matching and Pair Modeling29
This work proposes DiPair — a novel framework for distilling fast and accurate models on text pair tasks that is both highly scalable and offers improved quality-speed tradeoffs.
- Exploiting Unlabeled Data for Neural Grammatical Error Detection29
An attention-based neural network is introduced to capture long-distance dependencies that influence the word being detected and significantly outperforms SVM and convolutional networks with fixed-size context window.
- 26
- Downwash-aware Control Allocation for Over-actuated UAV Platforms26
This work proposes a novel control allocation framework that considers downwash effects and explores the entire allocation space for an optimal solution, which avoids down wash effects while providing high thrust efficiency within the hardware constraints.
- Breaching FedMD: Image Recovery via Paired-Logits Inversion Attack25
This study shows that a malicious server can inject a PLI (Paired-Logits Inversion) attack against FedMD and its variants by training an inversion neural network that exploits the confidence gap between the server and client models.
- UniSumm and SummZoo: Unified Model and Diverse Benchmark for Few-Shot Summarization24
Experimental results and analysis show that UniSumm outperforms strong baselines by a large margin across all sub-tasks in SummZoo under both automatic and human evaluations and achieves comparable results in human evaluation compared with a GPT-3.5 model.
- 23
- Bilingual Correspondence Recursive Autoencoder for Statistical Machine Translation23
This paper proposes a new neural network model called Bilingual Correspondence Recursive Autoencoder (BCorrRAE) to model bilingual phrases in translation and incorporates word alignments into BCorrRAE to allow it freely access bilingual constraints at different levels.
- Bilingual Correspondence Recursive Autoencoders for Statistical Machine Translation23
This paper proposes a new neural network model called Bilingual Correspondence Recursive Autoencoder (BCorrRAE) to model bilingual phrases in translation and incorporates word alignments into BCorrRAE to allow it freely access bilingual constraints at different levels.
- 22
- On the Language Coverage Bias for Neural Machine Translation22
Two simple and effective approaches to alleviate the language coverage bias problem are proposed through explicitly distinguishing between the source- and target-original training data, which consistently improve the performance over strong baselines on six WMT20 translation tasks.
- Deep Learning in Machine Translation22
This chapter introduces two broad categories of deep learning-based MT methods: component-wise deep learning for machine translation that leverages deep learning to improve the capacity of the main components of SMT such as translation models, reordering models, and language models; and end-to-end deep learning that uses neural networks to directly map between source and target languages based on the encoder–decoder framework.
- Assessing computational tools for predicting protein stability changes upon missense mutations using a new dataset21
A new dataset from three recently published databases, consisting of 4038 single‐point mutations, including over 1000 stabilizing mutations, is compiled using 27 computational methods, including the latest ones utilizing mega‐scale stability datasets and transfer learning.
- 21
- Black-Box Prompt Tuning With Subspace Learning19
This paper introduces Black-box prompt tuning with Subspace Learning (BSL) to enhance the versatility of black-box prompt tuning and proposes identifying such subspaces through meta-learning on a collection of similar source tasks.
- Integrating Spectral, Textural, and Morphological Data for Potato LAI Estimation from UAV Images19
The combination of different features enhanced the simulation of LAI for multiple growth stages of potato plants by improving the monitoring accuracy and the method presented in this study can provide important references for potato plant growth monitoring.
- Data augmentation for low‐resource languages NMT guided by constrained sampling19
This work tries to augment the corpus by introducing a constrained sampling method that consistently outperforms all the previous state‐of‐the‐art methods on both small and large‐scale corpora in eight language pairs from four corpora.
- 16
- Learning from Chinese-English Parallel Data for Chinese Tense Prediction16
The results demonstrate the effectiveness of the proposed learning framework that maps annotation from one language to another using parallel data and shows better performance using the proposed iterative bootstrapping learning method compared to using the original automatically created training data.
- Sub-sentence division for tree-based machine translation16
This work proposes a new method named sub-sentence division that reduces the decoding time and improves the translation quality for tree-based translation by dividing long sentences into several sub- Sentences by exploiting tree structures.
- Combining Multiple Alignments to Improve Machine Translation15
Experiments show that this approach not only improves the alignment quality, but also significantly improves translation performance by up to 1.96 BLEU points over single best alignments, and 1.28 points over merging rules extracted from multiple alignments individually.
- Weakly-Supervised Open-Retrieval Conversational Question Answering14
This work introduces a learned weak supervision approach that can identify a paraphrased span of the known answer in a passage of a passage in the open-retrieval ConvQA setting under a weak supervision setting.
- Improving Data Augmentation for Low-Resource NMT Guided by POS-Tagging and Paraphrase Embedding13
This work generates the fake monolingual corpus by replacing the main four POS-Tagging labels, such as noun, adjective, adverb, and verb, based on both the paraphrase table and their similarity, and exploits the ranking algorithm to choose highly similar words to reduce semantic errors.
- 13
- 12
- Diagnosing pathologic myopia by identifying morphologic patterns using ultra widefield images with deep learning11
This study aims to diagnose pathologic myopia by identifying clinically significant morphologic patterns, specifically posterior staphyloma and myopic maculopathy, by leveraging ultra-widefield (UWF) images that provide a broad retinal field of view by introducing RealMNet, an end-to-end lightweight framework designed to identify these challenging patterns.
- The LAIX Systems in the BEA-2019 GEC Shared Task10
This paper describes two GEC systems developed for the BEA-2019 GEC Shared Task, and uses ensemble systems to selectively combine the NMT models, the classification models, and some rules, and demonstrates that an ensemble solution can effectively improve GEC performance over single systems.
- 9
- 9
- 9
- 9
- 9
- 9
- 8
- 8
- 8
- 7
- 7
- Alignment-consistent recursive neural networks for bilingual phrase embeddings7
This paper considers word alignments as a semantic bridge between the source and target phrases, and proposes two neural networks based on the conventional recursive autocoder, which exploitword alignments to generate alignment-consistent bilingual phrase structures.
- 6
- Unsupervised Discriminative Induction of Synchronous Grammar for Machine Translation6
A global log-linear model for synchronous grammar induction, which is capable of incorporating arbitrary features, is presented and a novel and efficient cube pruning based synchronous parsing algo- rithm is proposed.
- 6
- 5
- 5
- Error Analysis of Uyghur Name Tagging: Language-specific Techniques and Remaining Challenges5
It is concluded that simply adopting a machine learning model which is proven successful for high-resource languages along with language-independent superficial features is unlikely to be effective for Uyghur, or low- resource languages in general.
- To Swap or Not to Swap? Exploiting Dependency Word Pairs for Reordering in Statistical Machine Translation5
A novel reordering approach utilizing sparse features based on dependency word pairs that captures whether two words, which are related by a dependency link in the source sentence dependency parse tree, follow the same order or are swapped in the translation output.
- 5
- 4
- 4
- 4
- 4
- 3
- The ICT Statistical Machine Translation Systems for IWSLT 20073
An overview of the ICT statistical machine translation systems for the evaluation campaign of the International Workshop on Spoken Language Translation (IWSLT) 2007 is given, which ranks 2 among the 15 primary results according to the official evaluation results.
- 2
- 2
- Multi-Zone Unit for Recurrent Neural Networks2
A new Multi-zone Unit (MZU) for RNNs is introduced, to design a transition function that is capable of modeling multiple space composition and experimental results demonstrate the superiority of the MZU.
- 2
- 2
- 2
- Dependency-Based Bracketing Transduction Grammar for Statistical Machine Translation2
A novel dependency-based bracketing transduction grammar for statistical machine translation, which converts a source sentence into a target dependency tree, and employs two different maximum entropy models to determine the reordering and combination of partial dependency structures.
- The ICT Statistical Machine Translation Systems for the IWSLT 20092
This paper describes the ICT Statistical Machine Translation systems that used in the evaluation campaign of the International Workshop on Spoken Language Translation (IWSLT) 2009 and develops a sentence-similarity based development set selection technique.
- 1
- 1
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
Publication data from OpenAlex, with missing venues and authors filled in from Crossref; citation counts are the higher of OpenAlex and Semantic Scholar, last synced 2026-10-16. One-sentence summaries under some papers are written by Semantic Scholar’s model. Citation counts may be lower than on Google Scholar, which indexes more sources.
Report an error
Wrong papers, two people merged into one, or a profile that should not be here? Tell us and we will fix or hide it.