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 32 for “"Model Alignment"”.
-
Robust Methods for Visual Tracking and Model Alignment
… stable structures, the parameterized shape model of objects can be usually built or learned from sample images, which allows us to perform more accurate tracking. We present a machine learning method to learn a scoring function without local extrema to guide the gradient descent/accent …
-
Methods for Language Model Alignment: Inference-Time Sampling and Preference Fine-Tuning
As large language models (LLMs) transition from research prototypes to widely deployed systems, aligning their outputs with human preferences and objectives has become a central challenge. While pretrained LLMs generate fluent text, they often fail to follow instructions, meet users’ objectives, or …
-
Topics in Statistical Machine Learning: Multitask Learning, Uncertainty Quantification, and Language Model Alignment
… uncertainty quantification, andlanguage model alignment. In the first chapter, we investigate the challenges and opportunities in learning from large, heterogeneousdatasets, where data is collected from multiple sources with sparse heterogeneity, meaning that the source/task-associated …
-
PREFERENCE LEARNING AT THE INTERFACE OF AI AND OPERATIONS
… develops a novel, tractable Mallows-type ranking model that enables preferences to be learned from ranked choices and supports the design of efficient feedback systems. Moving in the opposite direction, the second study brings choice modeling ideas from operations research to AI alignment. It …
-
Mathematical methods for 2D-3D cardiac image registration
… are used to validate the parametric model. Target registration error (TRE), Jaccard, and Dice indices are used to evaluate the algorithm and demonstrate the accuracy of the registration scheme on both simulated and clinical data. Registration with the affine model appeared to be more …
-
Aligning Language Models Using Multi-Objective Deep Reinforcement Learning
Large Language Models (LLMs) have been a significant landmark of Artificial Intelligence (AI) advancement. Aligning LLMs to be helpful and harmless is a booming trend in Natural Language Processing (NLP). One of the dominant alignment techniques is reinforcement learning from human feedback (RLHF). …
-
Exploring the Effects of Explainable Artificial Intelligence (XAI) Across AI Tools with Different Levels of Decision Complexity
Drawing on Mental Models Theory and Task Complexity Theory, this research examines how visual and textual Explainable Artificial Intelligence (XAI) explanations influence users’ trust and mental model clarity in AI recommendation systems. Using a 2 × 3 mixed factorial experiment, the study compared …
-
From Sustainability Ambition to Decision-Making in Practice
… requirements, internal priorities, business model alignment, and project-level conditions. Sustainability gains influence when it can be connected to customer value, competitiveness, regulation, risk reduction, or measurable outcomes such as certifications, climate targets, and KPIs. However, …
-
Cascaded refactoring for framework development and evolution
… extension. A framework is specified by a set of models: feature model, use case model, architectural model, design model, and source code. Framework refactoring is achieved by a set of refactorings cascaded from the feature model, to use case model, architectural model, design model, and source …
-
Investigation of a Validation Technique for the Registration of 3D Points Directly to Subjects with Pectus Excavatum
… scan acquisition, creation of the rib cage model from CT data, and an approach to rough model alignment prior to registration have been made. Future steps following this research include the investigation of accurate point identification on and data collection from a subject with PE.
-
LHCb spectrometer alignment and verification of its performance using the decay B [sub d exp 0] → K*⁰ J/ψ
… is crucial to this experiment. Detector alignment is an essential ingredient for obtaining the best possible measurement precision of a detector. The internal and relative alignment of the LHCb tracking subdetectors is performed using a novel method based on a Kalman filter track model. …
-
On the Resource Efficiency of Language Models
Large language models (LLMs) have revolutionized a wide range of natural language processing tasks. However, their practical utility faces resource challenges in two dimensions: data efficiency and model efficiency. For post-training, LLMs face data curation challenges where high-quality labeled …
-
Vancouver's living wage campaign : social movement unionism and identity construction
… within a social movement unionism organizing model. Based on findings from interviews with the outsourced workers, I recommend modifications to mobilization theory because within a social movement unionism model, campaign success depends, in part, on workers actively shaping the interpretive …
-
Learnable Interfaces - Leveraging Navigation by Design
… and the task domain when a conceptual model of the task domain is intentionally embedded. This can lead to more productive communication lower development and training costs, and improved task performance. The study targets the navigation interface as a gateway to exposing structural …
-
Managing Consistency of Business Process Models across Abstraction Levels
Process models support the transition from business requirements to IT implementations. An organization that adopts process modeling often maintain several co-existing models of the same business process. These models target different abstraction levels and stakeholder perspectives. Maintaining …
-
Trustworthy Autonomy Through Robust Control and Alignment
… reliable control through formal guarantees and alignment against adversarial manipulation. The first part of the dissertation focuses on provably stable, robust, and safe control for nonlinear systems using learning based methods. We introduce the first general framework for synthesizing neural …
-
Optimization methods for reinforcement learning: theory and applications
… simulated robotic tasks and a large language model alignment task, highlighting the effectiveness of our mirror descent-based approach for preference learning.
-
Measurement of Surface Heat Flux Due to Fin-Induced Shock-Boundary Layer Interactions
… (SBLI) at Texas AM University. Wind tunnel models were 3D printed from Formlabs Rigid 10K resin using a Formlabs 3L printer. To accurately quantify heating caused by fins, a new method was developed to convert infrared temperature measurements into heat flux using the full, 3D heat equation. …
-
Efficient Algorithm for Learning and Generation with Reinforcement Learning
… costs for multi-objective Pareto optimal bound alignment. This thesis explores these efficiency challenges through three distinct yet interconnected RL contexts, developing theoretically grounded algorithms that overcome traditional limitations while maintaining practical applicability across …
-
Efficient Algorithm for Learning and Generation with Reinforcement Learning
… costs for multi-objective Pareto optimal bound alignment. This thesis explores these efficiency challenges through three distinct yet interconnected RL contexts, developing theoretically grounded algorithms that overcome traditional limitations while maintaining practical applicability across …
Page 1 of 2