Multilingual Meta-Embeddings (MME) leverage information from pre-trained mono-lingual embeddings. The exploited multi-lingual information can be effectively used in the code-mixed domain. This paper maneuvers MME to p...
详细信息
ISBN:
(数字)9781665408370
ISBN:
(纸本)9781665408387;9781665408370
Multilingual Meta-Embeddings (MME) leverage information from pre-trained mono-lingual embeddings. The exploited multi-lingual information can be effectively used in the code-mixed domain. This paper maneuvers MME to propose a language identification mechanism for code-mixed text. empirical evaluation of conventional deep learning models in discrete, and permutated forms, GRU, LSTM, CRF, GRU with CRF, LSTM with CRF,LSTM with CNN, and GRU with CNN. Multi-Task Learning approach and evaluated our models on the Lince Hindi-English Code Mixed language Identification data set.
Hierarchical structures exist in both linguistics and naturallanguageprocessing (NLP) tasks. How to design RNNs to learn hierarchical representations of naturallanguages remains a long-standing challenge. In this p...
详细信息
ISBN:
(纸本)9780738133669
Hierarchical structures exist in both linguistics and naturallanguageprocessing (NLP) tasks. How to design RNNs to learn hierarchical representations of naturallanguages remains a long-standing challenge. In this paper, we define two different types of boundaries referred to as static and dynamic boundaries, respectively, and then use them to construct a multi-layer hierarchical structure for document classification tasks. In particular, we focus on a three-layer hierarchical structure with static word- and sentence- layers and a dynamic phrase-layer. LSTM cells and two boundary detectors are used to implement the proposed structure, and the resulting network is called the Recurrent Neural Network with Mixed Hierarchical Structures (MHS-RNN). We further add three layers of attention mechanisms to the MHS-RNN model. Incorporating attention mechanisms allows our model to use more important content to construct document representation and enhance its performance on document classification tasks. Experiments on five different datasets show that the proposed architecture outperforms previous methods on all the five tasks.
Active learning (AL) is a technique for reducing manual annotation effort during the annotation of training data for machine learning classifiers. For NLP tasks, pool-based and stream-based sampling techniques have be...
详细信息
End-to-end spoken language understanding (SLU) remains elusive even with current large pretrained language models on text and speech, especially in multilingual cases. Machine translation has been established as a pow...
详细信息
ISBN:
(纸本)9798891760615
End-to-end spoken language understanding (SLU) remains elusive even with current large pretrained language models on text and speech, especially in multilingual cases. Machine translation has been established as a powerful pretraining objective on text as it enables the model to capture high-level semantics of the input utterance and associations between different languages, which is desired for speech models that work on lower-level acoustic frames. Motivated particularly by the task of cross-lingual SLU, we demonstrate that the task of speech translation (ST) is a good means of pretraining speech models for end-to-end SLU on both intra- and cross-lingual scenarios. By introducing ST, our models reach higher performance over baselines on monolingual and multilingual intent classification as well as spoken question answering using SLURP, MINDS-14, and NMSQA benchmarks. To verify the effectiveness of our methods, we also create new benchmark datasets from both synthetic and real sources, for speech summarization and low-resource/zero-shot transfer from English to French or Spanish. We further show the value of preserving knowledge for the ST pretraining task for better downstream performance, possibly using Bayesian transfer regularizers.
naturallanguage inference (NLI) aims to infer the relationship between two texts: premise and hypothesis. However, many existing methods overlook the problem of overestimation of model performance due to superficial ...
详细信息
ISBN:
(纸本)9798350344868;9798350344851
naturallanguage inference (NLI) aims to infer the relationship between two texts: premise and hypothesis. However, many existing methods overlook the problem of overestimation of model performance due to superficial correlation biases in NLI datasets. We study this problem and find that most current models have taken NLI as one of the text-matching tasks, which ignores the asymmetry of the premise and hypothesis of NLI. Therefore, we propose a simple and effective augmentation method, Inversive-Reasoning Augmentation (IRA), to remove the superficial correlation bias. After training the different NLI models with our IRA-augmented data based on two widely-used NLI datasets, we observemore fair evaluation results of the performance and robustness of the various NLI models.
We investigate the usefulness of evolutionary algorithms in three incarnations of the problem of feature relevance assignment in memory-based languageprocessing (MBLP): feature weighting, feature ordering and feature...
详细信息
A key component of fact verification is the evidence retrieval, often from multiple documents. Recent approaches use dense representations and condition the retrieval of each document on the previously retrieved ones....
详细信息
Foundation models and vision-language pretraining have notably advanced Vision language Models (VLMs), enabling multimodal processing of visual and linguistic data. However, their performance has been typically assess...
详细信息
Large language Models (LLMs) have shown promising performance in text-to-SQL, which involves translating naturallanguage questions into SQL queries. However, current text-to-SQL LLMs are computationally expensive and...
详细信息
Over the last three years, work of an empirical nature has been carried out on the design of a natural-language question-answering system. In the present paper, the parser-analyzer of version 3.0 of the QUANSY (gt/est...
详细信息
暂无评论