Global ETD Search

Search theses and dissertations gathered from participating repositories worldwide. Every result links back to the library that holds it. No account is needed.

Results

Showing 1 to 20 of 42 for “"sentence embeddings"”.

  1. Exploring sentence embeddings for better argument relation classification

    … argument relation classification by augmenting sentence embeddings and leveraging Large Language Models (LLMs). Argument relation classification, a subtask of argument mining, involves predicting relationships, such as support or attack, between argumentative components in unstructured text. …

    soton Download PDF for Exploring sentence embeddings for better argument relation classification (opens in a new tab)

  2. Deep Latent-Variable Models for Natural Language Understanding and Generation

    … that can be helpful for tasks such as sentence classification, natural language inference, question answering, etc. Firstly, I will propose a variational autoencoder based on textual data to digest unlabeled information. To alleviate the observed posterior collapse issue, a …

    duke Repository record for Deep Latent-Variable Models for Natural Language Understanding and Generation (opens in a new tab)

  3. Melhorando o rotulamento de papel retórico legal através de dados adicionais e uso eficiente de modelos transformadores

    … (RRL), a task that assigns semantic functions to sentences in legal documents. Legal RRL is a relevant task because it finds information that is useful both by itself and for downstream tasks such as legal summarization and case law retrieval. There are several factors that make legal RRL a …

    brazil-ufrn Repository record for Melhorando o rotulamento de papel retórico legal através de dados adicionais e uso eficiente de modelos transformadores (opens in a new tab)

  4. An Unsupervised Approach to Relatedness Analysis of Legal Language

    Learning distributed representations of sentences and analyzing semantic similarity between sentences is one of the essential works in the field of Natural Language Processing. In the domain of legal language, the future of Artificial Intelligence-related legal-tech applications is very promising. …

    waterloo Repository record for An Unsupervised Approach to Relatedness Analysis of Legal Language (opens in a new tab)

  5. Contextual topics: advancing text segmentation through pre-trained models and contextual keywords

    … of BERT. It is also found that infusing sentence embeddings with unsupervised features, such as the ones gathered from Latent Dirichlet Allocation (LDA), provides comparable results to current state-of-the-art (SOTA) TS systems. In addition to this, unsupervised features derived from LDA …

    uoit Repository record for Contextual topics: advancing text segmentation through pre-trained models and contextual keywords (opens in a new tab)

  6. The Fusion of Multilingual Semantic Search and Large Language Models: A New Paradigm for Enhanced Topic Exploration and Contextual Search

    … suggestion. By conducting experiments with Sentence Embeddings, Large Language Model (LLM), and Approximate Nearest Neighbours (ANN), this thesis improves news article search systems. Results show that the method improves the quality of the search results for the queries to return more …

    carleton Repository record for The Fusion of Multilingual Semantic Search and Large Language Models: A New Paradigm for Enhanced Topic Exploration and Contextual Search (opens in a new tab)

  7. Pairwise embedding for event coreference resolution

    … These features include but are not limited to sentence embeddings, trigger embeddings, and argument type representations. The extracted features are then used to fine-tune a pre-trained language model, BERT (Bidirectional Encoder Representations from Transformers), to perform event coreference …

    uiuc Repository record for Pairwise embedding for event coreference resolution (opens in a new tab)

  8. Detecting the Intent of Email Using Embeddings, Deep Learning and Transfer Learning

    … the classification phases by using our own word embeddings, trained on the entire Enron Email dataset, to represent the input. Then, we use a convolutional layer to capture local tri-gram features, followed by an LSTM layer to consider the meaning of a given feature (trigrams) concerning some …

    south-carolina Repository record for Detecting the Intent of Email Using Embeddings, Deep Learning and Transfer Learning (opens in a new tab)

  9. Automatic Summarization of Financial Reports

    … with various non-parametric methods to generate sentence vectors which can facilitate the efficient selection of important sentences in the document as part of the summarization algorithm. We illustrate the superiority of a financially tuned language model. Further, it is shown that simple …

    regina Repository record for Automatic Summarization of Financial Reports (opens in a new tab) Download PDF for Automatic Summarization of Financial Reports (opens in a new tab)

  10. Predicting the Emotional Intensity of Tweets

    … also use a form of transfer learning – word and sentence embeddings extracted from neural networks trained on large corpora. The estimation of emotional intensity is a regression task and we use linear and tree-based models for this task. We compare the results of these individual models as well …

    rit Repository record for Predicting the Emotional Intensity of Tweets (opens in a new tab)

  11. From invoices to insights : feature engineering strategies for multilingual product classification in medical device procurement

    … descriptions. A feature ladder shows that sentence embeddings combined with vendor and unit cost already reach a weighted F1 of 0.950 at 4-digit ISO (five-country mean), and a single pooled pan-Nordic model raises this to 0.959; leave-one-country-out transfer retains 86–100% of …

    reykjavik Repository record for From invoices to insights : feature engineering strategies for multilingual product classification in medical device procurement (opens in a new tab)

  12. Producing synthetic instances for textual classification and natural language inference

    … on deduction – a model is presented to a pair of sentences (premise and hypothesis) and classifies the relationship between their meanings. Training models with NLI datasets is key for semantic systems. NLI datasets are also used to train sentence-transformer (ST) models, which use Siamese …

    brazil-ufrgs Repository record for Producing synthetic instances for textual classification and natural language inference (opens in a new tab)

  13. Producing synthetic instances for textual classification and natural language inference

    … on deduction – a model is presented to a pair of sentences (premise and hypothesis) and classifies the relationship between their meanings. Training models with NLI datasets is key for semantic systems. NLI datasets are also used to train sentence-transformer (ST) models, which use Siamese …

    brazil-ufrgs Repository record for Producing synthetic instances for textual classification and natural language inference (opens in a new tab)

  14. Weakly supervised learning from referring expression: Challenge and directions

    … expression task is performed associated with a sentence phrase, e.g. “the dude on the dolphin”. Previous approaches use LSTM and fully convolutional network and have fairly good results under fully supervised setting. However, the fully supervised setting is limited by manual labeling of …

    uiuc Repository record for Weakly supervised learning from referring expression: Challenge and directions (opens in a new tab)

  15. Smart transfers: challenges and opportunities in boosting low-resource language models with high-resource language power

    … features and biases. The second stream develops sentence Transformers, which generate sentence embeddings by mean-pooling contextualized word embeddings. However, these embeddings often struggle to capture sentence similarities effectively. To address this, we fine-tuned an English sentence …

    colostate Repository record for Smart transfers: challenges and opportunities in boosting low-resource language models with high-resource language power (opens in a new tab)

  16. Clustering and Sentiment Analysis of Customer NVH Feedback in the Automotive Domain A Machine Learning Pipeline to Facilitate Extraction of Relevant Information from Large-Scale Textual Data

    … feedback. The proposed pipeline combined sentence embeddings, dimensionality reduction, and clustering to group semantically similar feedback. Cluster labels were automatically generated using a large language model and manually refined when necessary. Both sentencebased and aspect-based …

    chalmers Repository record for Clustering and Sentiment Analysis of Customer NVH Feedback in the Automotive Domain A Machine Learning Pipeline to Facilitate Extraction of Relevant Information from Large-Scale Textual Data (opens in a new tab)

  17. Towards Explainable Event Detection and Extraction

    … events of interest and from those extracting the sentences that might contain the event related information. This task is challenging because it is easier to obtain labels at the document level than finegrained annotations at the sentence level. Current approaches for this task are suboptimal …

    vt Repository record for Towards Explainable Event Detection and Extraction (opens in a new tab)

  18. On the Ethics and Linguistic Impacts of Using the Bible as Training Data for Yucatec Maya-to-Spanish Machine Translation

    … a semantic similarity analysis using BETO sentence embeddings, and a Bible n-gram contamination analysis using log-likelihood ratio statistics. Results show that increasing the proportion of Bible training data consistently degraded translation quality across both models. For …

    washington Repository record for On the Ethics and Linguistic Impacts of Using the Bible as Training Data for Yucatec Maya-to-Spanish Machine Translation (opens in a new tab)

  19. Sentence Similarity and Machine Translation

    … translation (NMT) systems encode an input sentence into an intermediate representation and then decode that representation into the output sentence. Translation requires deep understanding of language; as a result, NMT models trained on large amounts of data develop a semantically rich …

    jhu Repository record for Sentence Similarity and Machine Translation (opens in a new tab)

  20. Multiple Choice Question Answering using a Large Corpus of Information

    … 2 I review popular algorithms to create word and sentence embeddings given the surrounding context. These embeddings are a numerical representation of the language data that can be used in downhill models such as logistic regression. In Chapter 3 I present a novel method to create a domain specific …

    umn Repository record for Multiple Choice Question Answering using a Large Corpus of Information (opens in a new tab)

Page 1 of 3