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 10 of 10 for “"pretrained transformers"”.

  1. Pretrained Transformers for Efficient and Robust Information Retrieval

    Pretrained transformers have significantly advanced the field of information retrieval (IR) since the introduction of BERT. Through unsupervised pretraining followed by task-aware finetuning, the encoder-only BERT effectively captures the complex semantics of texts and overcomes the limitations of …

    waterloo Repository record for Pretrained Transformers for Efficient and Robust Information Retrieval (opens in a new tab)

  2. Self-supervised text sentiment transfer with rationale predictions and pretrained transformers

    … identification method. We also experiment with a pretrained encoder-decoder Transformer model, known as BART, as a method for improving upon previous sentiment transfer results. This pretrained model is fine-tuned first in an unsupervised manner as a denoising autoencoder to reconstruct sentences …

    cape-town Repository record for Self-supervised text sentiment transfer with rationale predictions and pretrained transformers (opens in a new tab)

  3. Future Sight: Dynamic Story Generation with Large Pretrained Language Models

    … inference. We build an interactive system using pretrained transformers finetuned on a novel objective to temporally interpolate between a story context c and a future plot event f. At inference time, users can suggest future plot events along with a distance, in sentences, to coerce a …

    waterloo Repository record for Future Sight: Dynamic Story Generation with Large Pretrained Language Models (opens in a new tab)

  4. Improving Relation Extraction from Unstructured Genealogical Texts Using Fine-Tuned Transformers

    … makes use of recently developed and massively pretrained transformers and fine-tuning techniques to take advantage of these deep neural models’ inherent understanding of English syntax and semantics for classification. With only a fraction of labeled data typically needed to train large models, …

    calpoly Repository record for Improving Relation Extraction from Unstructured Genealogical Texts Using Fine-Tuned Transformers (opens in a new tab)

  5. SCALING PRE-TRAINING DATA & LANGUAGE MODELS FOR AFRICAN LANGUAGES

    Recent advancements in language models, particularly for high-resource languages, have not been paralleled in low-resource languages spoken across Africa. This thesis addresses this gap by scaling pre-training data and developing improved language models for African languages. We introduce Wura, a …

    waterloo Repository record for SCALING PRE-TRAINING DATA & LANGUAGE MODELS FOR AFRICAN LANGUAGES (opens in a new tab)

  6. Multilingual Financial Text Summarisation

    … text in multiple languages. Using advanced pretrained transformers, our system can accurately identify and extract essential information from financial documents and generate extractive and abstractive summaries in various languages.

    lancaster Download PDF for Multilingual Financial Text Summarisation (opens in a new tab)

  7. The power of adaptivity in multi-task representation learning

    … Second, we consider multi-task learning with transformers in the setting of in-context learning (ICL). ICL entails solving a new task simply by running a forward pass on a prompt consisting of an instruction and examples of how to solve the task, with no model updates involved. Pretrained …

    texas Repository record for The power of adaptivity in multi-task representation learning (opens in a new tab)

  8. Development and Evaluation of Vision-Language Models for Ergonomic Physical Exposure Assessment in Occupational Manual Material Handling

    Work-related musculoskeletal disorders remain a leading cause of lost workdays and substantial costs for workers and employers, and manual material handling (MMH) tasks are a major contributor to the physical exposures that drive this risk. Existing physical exposure assessment methods, including …

    vt Repository record for Development and Evaluation of Vision-Language Models for Ergonomic Physical Exposure Assessment in Occupational Manual Material Handling (opens in a new tab)

  9. Democratizing and Modernizing Information Access: From Open Rerankers to Scalable RAG Evaluation

    Modern information access increasingly relies on complex pipelines involving large language models (LLMs), fundamentally changing how users interact with information, from sophisticated multi-stage retrieval pipelines to end-to-end retrieval-augmented generation (RAG) systems. While these …

    waterloo Repository record for Democratizing and Modernizing Information Access: From Open Rerankers to Scalable RAG Evaluation (opens in a new tab)

  10. Benchmarks, Data, and Evaluation for Robust Retrieval and Retrieval-Augmented Generation on Heterogeneous Domains and Languages

    … a retrieval system. Similarly, with the onset of pretrained transformers, retrieval systems have demonstrated strong downstream accuracy in the in-domain setting. However, their robustness in the zero-shot scenario remains limited, suggesting that these assumptions do not consistently hold in …

    waterloo Repository record for Benchmarks, Data, and Evaluation for Robust Retrieval and Retrieval-Augmented Generation on Heterogeneous Domains and Languages (opens in a new tab)