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 20 for “"PROMPT LEARNING"”.

  1. Explainable Prompt Learning for Movie Review Sentiment Analysis

    … for applying these models to various tasks is prompting, where we present a model with a prompt that guides it towards solving a particular task. This approach has achieved success on a variety of tasks and has led to the concept of prompt learning, where we fine-tune the language models on the …

    ottawa-retro Repository record for Explainable Prompt Learning for Movie Review Sentiment Analysis (opens in a new tab)

  2. ENHANCING ZERO-SHOT SEGMENTATION WITH DOMAIN-ADAPTIVE PROMPT LEARNING

    … knowledge and often require extensive learnable prompt engineering. This project aims to overcome these limitations by introducing domain-adaptive prompt learning into a CLIP-based ZSSeg pipeline. Our approach enhances CLIP-based segmentation with learnable domain-agnostic and domain-specific …

    csusb Repository record for ENHANCING ZERO-SHOT SEGMENTATION WITH DOMAIN-ADAPTIVE PROMPT LEARNING (opens in a new tab)

  3. Enhancing Low-Resource Language Performance in Multilingual Large Language Models

    … developed knowledge distillation and strategic prompt-learning, and attention alignment methods to improve the representation capabilities of large language models for low-resource language, and then enhanced their performance in downstream tasks. In our first study, we developed a new knowledge …

    clemson Repository record for Enhancing Low-Resource Language Performance in Multilingual Large Language Models (opens in a new tab)

  4. Generalizable Adaptation for Vision-Language Models

    … have recently emerged as powerful frameworks for learning joint representations across visual and textual modalities. These models enable a wide range of applications, including visual recognition, multimodal reasoning, and visual question answering. However, adapting large pre-trained VLMs to …

    clemson Repository record for Generalizable Adaptation for Vision-Language Models (opens in a new tab)

  5. Annotation-efficient and reliable medical image segmentation

    … planning, and disease monitoring. While deep learning based algorithms have delivered state-of-the-art segmentation performance, their integration into the clinical workflow is hindered by two major challenges: the high cost of acquiring large annotated datasets and the lack of guarantees on …

    ets-quebec Download PDF for Annotation-efficient and reliable medical image segmentation (opens in a new tab)

  6. Towards Intelligent Conversational Assistants: Enhancing Task-Oriented Dialogue Systems with Knowledge Integration

    … dialogue state tracking, schema graph-guide prompt learning, dialogue-level modeling using dialogue acts and a soft mixture-of-experts framework for TOD systems. These innovations enhance dialogue management accuracy, contextual understanding, and system modularity. Additionally, a …

    gatech Repository record for Towards Intelligent Conversational Assistants: Enhancing Task-Oriented Dialogue Systems with Knowledge Integration (opens in a new tab)

  7. Efficient large language models for the NHS and psychiatry

    … models. Efficient adaptation methods, such as prompt learning, were devel- oped and employed, reducing computational and storage requirements by up to 98% while maintaining state-of-the-art performance on several clinical NLP tasks. The bespoke NHS LLMs and efficient adaptation methods were …

    oxford Repository record for Efficient large language models for the NHS and psychiatry (opens in a new tab)

  8. How Primary School Students' Arguments Develop: Taking Initial Steps in a Deductive Discourse

    … proof, what might constitute smaller steps of learning and how young students can be supported to take these steps. A commognitive perspective is taken to examine these gaps in the research. The research focuses on Year 4 students’ discourses about odd and even numbers and their substantiations …

    auckland-ms Repository record for How Primary School Students' Arguments Develop: Taking Initial Steps in a Deductive Discourse (opens in a new tab)

  9. Distributional robustness: towards challenging and real-world settings of distribution shift

    … data distribution shifts for any machine learning application. As machine learning models are notoriously not robust against such distribution shifts, it is of our interest to develop techniques to deal with this problem. <strong>Hence, this thesis investigates the issue of distribution …

    oxford Repository record for Distributional robustness: towards challenging and real-world settings of distribution shift (opens in a new tab)

  10. Exploring the Paradoxical Role and Experience of Fallback in Developmental Theory

    … to an earlier stage of meaning-making may prompt learning that can lead to growth. The purpose of this study, which used grounded theory methodology, was to examine the phenomenon of fallback through twenty-seven interviews with six key thinkers in the field of adult and leader development: …

    usd-thes Repository record for Exploring the Paradoxical Role and Experience of Fallback in Developmental Theory (opens in a new tab)

  11. Towards Trustworthy Artificial Intelligence through Causal Inference

    … rely on correlation-based statistical learning, which, while effective in many applications, struggles to capture the causal mechanisms underlying real-world data. This limitation undermines the ability of AI systems to provide reliable reasoning, particularly in dynamic or complex …

    unsw Repository record for Towards Trustworthy Artificial Intelligence through Causal Inference (opens in a new tab)

  12. A Multimodal Framework for Automated Content Moderation of Children's Videos

    … techniques on the MOB benchmark. We perform prompt engineering and an in-depth analysis of how context-specific language prompts affect the content moderation performance of different CLIP (Contrastive Language-Image Pre-training) variants. This dissertation introduces new benchmark natural …

    ucf Repository record for A Multimodal Framework for Automated Content Moderation of Children's Videos (opens in a new tab)

  13. Towards Generalizable Representation Learning Across Domains

    Despite remarkable progress in deep learning, a major challenge remains: machine learning models often struggle to generalize to unseen domains under distribution shift. In real-world settings, data often differ from training conditions due to changes in lighting, sensor type, image resolution, and …

    clemson Repository record for Towards Generalizable Representation Learning Across Domains (opens in a new tab)

  14. Adaptive and Generalizable Vision-Language Models

    … inference. In this work, we introduce a domain prompt fusion framework aimed at improving the generalization capability of CLIP-based models under domain shift. Our approach integrates three core components: a dual-part soft prompt (comprising domain-agnostic and domain-specific prompts), a …

    chalmers Repository record for Adaptive and Generalizable Vision-Language Models (opens in a new tab)

  15. Efficient deep learning for image and video understanding

    … significant challenges remain in scaling up deep learning by considering the limited model/computation/data resources in real applications. In this thesis, we focus on these three major challenges: 1) memory efficiency, 2) training data efficiency, and 3) computation efficiency in popular computer …

    bu Repository record for Efficient deep learning for image and video understanding (opens in a new tab)

  16. Towards more flexible and efficient RGBD object tracking

    … cost problem in RGBD tracking by applying a prompt learning paradigm. With the proposed cross-modal prompts, both the large-scale RGB knowledge from pre-trained large models and complementary information from depth sensors can be well explored. Moreover, the prompting framework is effective …

    birmingham Download PDF for Towards more flexible and efficient RGBD object tracking (opens in a new tab)

  17. Artificial Intelligence Generated Content on Spatio-Temporal Modeling

    … accurate prediction. To address this, we propose PromptFDDM, which utilizes future motion data to enhance spatio-temporal modeling. The training process consists of two stages. In the first stage, real future motion data guides the Spatio-Temporal Extractor Network (STEN ), a framework designed to …

    anu Repository record for Artificial Intelligence Generated Content on Spatio-Temporal Modeling (opens in a new tab)

  18. Enhancing Text Representation by Minimizing Bias for Downstream Natural Language Processing Tasks

    … utilizing a novel self-supervised contrastive learning method. This model is specifically tuned to enhance the compatibility and effectiveness of MDERank in KPE tasks, addressing the discrepancies in representation that commonly arise in longer texts. The second study addresses frequency bias …

    unsw Repository record for Enhancing Text Representation by Minimizing Bias for Downstream Natural Language Processing Tasks (opens in a new tab)

  19. Constraint-based calibration for reliable deep learning models

    … CLIP adaptation strategies, such as Adapters, Prompt Learning, and Test-Time Prompt Tuning, substantially degrade the calibration capabilities of the zero-shot baseline in the presence of distributional drift. For these adaptation strategies, we expose that the underlying cause of …

    ets-quebec Download PDF for Constraint-based calibration for reliable deep learning models (opens in a new tab)