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 29 for “"In-context learning"”.

  1. Detecting Zero-Day Attacks in IEC-61850 based Digital Substations via In-Context Learning

    … on the electrical power grids have been increasing every year. In this thesis, we address the critical challenge of detecting novel/zero-day attacks in digital substations that employ the IEC-61850 communication protocol. While many heuristic and ML-based methods have been proposed for …

    vt Repository record for Detecting Zero-Day Attacks in IEC-61850 based Digital Substations via In-Context Learning (opens in a new tab)

  2. Study of Pretraining Bias and Frequencies

    Usage of language models in an in-context learning environment has been adapted for a wide range of tasks. Recent works have showcased the impact of pretraining data on the in-context performance of language models. In this work, we experiment with numbers having high and low frequencies in the …

    vt Repository record for Study of Pretraining Bias and Frequencies (opens in a new tab)

  3. Transformers as Empirical Bayes Estimators The Poisson Model

    We study the ability of transformers to perform In Context Learning (ICL) in the setting of Empirical Bayes for the Poison Model. On the theoretical side, we demonstrate the expressibility of transformers by formulating a way to approximate the Robbins estimator, the first empirical Bayes estimator …

    mit Repository record for Transformers as Empirical Bayes Estimators The Poisson Model (opens in a new tab)

  4. Learning to Segment Unseen Tasks In-Context

    While deep learning models have become the predominant method for medical image segmentation, they are typically incapable of generalizing to new segmentation tasks---involving new anatomies, image modalities, or labels. For a new segmentation task, researchers will often have to prepare new …

    mit Repository record for Learning to Segment Unseen Tasks In-Context (opens in a new tab)

  5. Clinical Text De-identification Using Large Language Models: Insights from Organ Procurement Data

    … a novel approach to the de-identification of clinical notes from Organ Procurement Organization (OPO) records, leveraging advanced natural language processing (NLP) methodologies. Specifically, we employ in-context learning using large language models (LLMs) to effectively identify and remove …

    mit Repository record for Clinical Text De-identification Using Large Language Models: Insights from Organ Procurement Data (opens in a new tab)

  6. Constrained and High-dimensional Bayesian Optimization with Transformers

    … algorithms that address critical limitations in handling constraints and high-dimensional spaces. First, we introduce a constraint-handling framework leveraging Prior-data Fitted Networks (PFNs), a foundation transformer model that evaluates objectives and constraints simultaneously in a …

    mit Repository record for Constrained and High-dimensional Bayesian Optimization with Transformers (opens in a new tab)

  7. Generalizing Under Data Scarcity. Enhancing the representation capability from few samples.

    The widespread adoption of deep learning in both research and industrial contexts has revealed a central limitation: many real-world applications lack large, diverse, and reliably labeled datasets. This challenge is particularly evident in domains where data acquisition is costly, error-prone, or …

    trento Repository record for Generalizing Under Data Scarcity. Enhancing the representation capability from few samples. (opens in a new tab)

  8. Leveraging Large Language Models (LLMs) for Automated Extraction and Processing of Complex Ordering Forms

    Data extraction from business documents is a critical but under-exploited area capable of unlocking significant value from vast document archives. Traditional methods relying on manual intervention or outsourcing are inefficient, error-prone, and costly, and commercial Deep Learning-based and OCR …

    mit Repository record for Leveraging Large Language Models (LLMs) for Automated Extraction and Processing of Complex Ordering Forms (opens in a new tab)

  9. Rethinking the Evaluation of Compositional Reasoning for Modern VLMs

    Recent advancements in modern Vision-Language Models (VLMs), comprising a visual encoder coupled with a Large Language Model (LLM) decoder, have demonstrated remarkable proficiency in Compositional Reasoning (CR). CR entails grasping the significance of attributes, relations, and word order. This …

    mit Repository record for Rethinking the Evaluation of Compositional Reasoning for Modern VLMs (opens in a new tab)

  10. New Approaches to Synthetic Tabular Data Generation

    Synthetic data generation, while already becoming well-known as part of Generative AI (GenAI), has been primarily focused on images, voice, and text, which mostly have homogeneous data formats. This dissertation focuses on the modeling and generation of synthetic tables, which involve a range of …

    vt Repository record for New Approaches to Synthetic Tabular Data Generation (opens in a new tab)

  11. Mixed-Variable Bayesian Optimization using Prior-Data Fitted Networks

    … (BO) is a powerful framework for optimizing expensive blackbox functions, widely used in domains such as materials science, engineering design, and hyperparameter tuning. Traditional BO relies on Gaussian processes (GPs) as surrogate models, but GPs face limitations in flexibility and …

    mit Repository record for Mixed-Variable Bayesian Optimization using Prior-Data Fitted Networks (opens in a new tab)

  12. Making Sense of Training Large AI Models

    … applications of optimization is the training of large AI models. But currently such models are trained with ad-hoc heuristics at a very large computational cost, mainly due to lack of understanding of their working mechanisms. In this thesis, we conduct a systematic study of large-model …

    mit Repository record for Making Sense of Training Large AI Models (opens in a new tab)

  13. LLM-Supported Natural Language to Bash Translation

    The Bourne-Again Shell (Bash) command-line interface for Linux systems has complex syntax and requires extensive specialized knowledge. Using the natural language to Bash command (NL2SH) translation capabilities of large language models (LLMs) for command composition alleviates these issues. …

    mit Repository record for LLM-Supported Natural Language to Bash Translation (opens in a new tab)

  14. Implicit capabilities of language models

    … models: abilities that emerge from standard training without the models being directly trained to possess them. As AI systems become more powerful, understanding such capabilities matters for anticipating these systems' behaviour and designing safety interventions. We present two such …

    cambridge Repository record for Implicit capabilities of language models (opens in a new tab)

  15. Meta-Learning Exploration Strategies with Decision Transformers

    The problem of pure exploration in sequential decision-making is to identify strategies for efficiently gathering information to uncover hidden properties of an environment. This challenge arises in many practical domains, including clinical diagnostics, recommender systems, and educational …

    mit Repository record for Meta-Learning Exploration Strategies with Decision Transformers (opens in a new tab)

  16. MOBLLM: Model Building LLMs via Symbolic Regression and Experimental Design

    … carry out more and more complex tasks every day, including those that require a high level of formal/mathematical reasoning at human or superhuman levels. In particular, their in-context learning capabilities and the domain-specific knowledge they have via their vast pretraining corpus, as well as …

    mit Repository record for MOBLLM: Model Building LLMs via Symbolic Regression and Experimental Design (opens in a new tab)

  17. Inference-Time Learning Algorithms of Language Models

    … models (LMs) can perform complex tasks through in-context learning (ICL)—they can adapt to a task via examples provided in their input without any parameter updates. However, fundamental questions remain about when this adaptation works, what algorithms underlie it, and how to improve it. This …

    mit Repository record for Inference-Time Learning Algorithms of Language Models (opens in a new tab)

  18. Learning Reconfigurable Vision Models

    Over the past decade, deep learning methods have emerged as the predominant approach in a wide variety of fields, such as computer vision, natural language processing, and speech recognition. However, these models have also been notorious for their high computational costs and substantial data …

    mit Repository record for Learning Reconfigurable Vision Models (opens in a new tab)

  19. Practical Considerations For the Deployment of Clinical NLP Systems

    Although recent advances in scaling large language models (LLMs) have resulted in improvements on many NLP tasks, it remains unclear whether these models trained primarily with general web text are the right tool in highly specialized, safety critical domains such as healthcare. A healthcare system …

    mit Repository record for Practical Considerations For the Deployment of Clinical NLP Systems (opens in a new tab)

  20. Exploring screen summarization with large language and multimodal models

    Mobile UIs are inherently multimodal. They can be represented by both visual representations, (screenshots) and structural metadata (view hierarchies). The image modality is rich and can contain information such as images, colors and positional information, while the view-hierarchies represent a …

    uiuc Repository record for Exploring screen summarization with large language and multimodal models (opens in a new tab)

Page 1 of 2