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 21 for “"Small Language Models"”.
-
Enabling Small Language Models as Efficient and Capable Agents
Most agentic systems today are built around large language models (LLMs) accessed through proprietary APIs (for example, GPT, Claude, and Gemini), which raises concerns about cost, latency, and privacy. This thesis argues that small language models (SLMs), typically under 30 billion parameters, can …
-
Reasoning beyond scale: Structured inference for small language models
Submission published under a 24 month embargo labeled 'U of I Access', the embargo will last until 2027-05-01
-
PhishGuard: Transparent Email Security Using A Dual Model Architecture
… privacy-aware architecture that combines natural language processing (NLP) modules with small language models (SLMs) for explainable and efficient phishing detection. The system adopts to use a dual architecture where lightweight NLP components extract syntactic and semantic features, while the …
-
Holistic Diagnosis via Multimodal Foundation Models
… and sound decisions. With the recent success of small and large language models, it is natural to try and incorporate them with multimodal capabilities in the healtcare domain. This thesis seeks to investigate how well small language models can perform on predictive tasks in healthcare using …
-
Causal Structure Learning for Robust and Explainable Artificial Intelligence
… develops an agentic framework using open-weight small language models that combines statistical evidence with agents for orienting, augmenting, and critiquing causal graphs. The framework is then extended with a causal story agent that reasons about the data-generating process to construct more …
-
Generalization and Fairness Optimization in Pretrained Language Models
… the generalization challenge in pretrained language models (PLMs), which affects their performance on diverse linguistic data beyond their training scope. Improving PLMs' adaptability to out-of-distribution (OOD) data is essential for their reliability and practical utility in real-world …
-
Study of Output and Behavior of LLMs Using Confidence Framing in Prompt Engineering
… prompt engineering is pivotal for shaping Large Language Model (LLM) outputs, the impact of confidence framing on behavioral calibration remains underexplored. This study investigates the ways in which psychological framing, utilizing techniques such as capability praise, role amplification, and …
-
Automated Methods to Evaluate the Quality of Distractors in College-Level MCQS
… the extent to which state-of-the-art large language models (LLMs) correlate with human rankings. Finally, we introduce a novel technique, "SLMs as Examinees" wherein multiple small language models (SLMs), trained to reflect varying levels of subject expertise, simulate human test-takers. …
-
Effective Automation of Black-Box Testing for REST APIs with Machine Learning and Language Models
… by incorporating machine learning and natural language processing (NLP) techniques to enhance automated test generation and execution. Through an empirical evaluation of ten state-of-the-art black-box testing tools across twenty RESTful web services, this research identified three key …
-
Tailoring large language models for zero-shot relation extraction
Submission original under an indefinite embargo labeled 'Open Access'. The submission was exported from vireo on 2025-10-19 without embargo terms
-
MEDICAL CLASSIFICATION THROUGH MACHINE LEARNING, DOMAIN-SPECIFIC LANGUAGE MODELS AND GENERATIVE AI TECHNIQUES
Advancing reliable machine learning and natural language processing methods for classification under data constrained conditions remains a central challenge in computer science. This dissertation addresses core problems related to data scarcity, class imbalance, heterogeneous tabular features, and …
-
Towards automated programming feedback with open-weight language models
… and nuanced guidance. Recent advances in language model (LM) research offer promising new avenues to address this gap. Large language models (LLMs) in particular have demonstrated strong capabilities for generating educational feedback. Much of the work in this space has relied on …
-
Adapting Transformers for Structured Data Domains
… domains beyond their traditional use in natural language processing (NLP). We revisit key elements of the transformer framework - including input representations, attention formulations, auxiliary tasks, prediction layers and loss functions - and adapt them to better suit the structure and …
-
Statistical Methods toward Trustworthy AI: From Diagnosis to Controllability and Societal Impact
… characterize and detect sources of bias in models ranging from generalized linear models to attention-based architectures. Part II (Control) introduces novel methods for controllable generation across different model scales and modalities. Chapter 3 develops an unsupervised, inference-time …
-
Advancing Information Extraction with Large Language Models: The Role of Structured Understanding in Knowledge Management and AI Safety
… scarcity of linguistic resources for non-English languages. This thesis investigates how Large Language Models (LLMs) can overcome these limitations and, conversely, how IE can enhance their reliability and safety. It examines the symbiotic relationship between structured knowledge extraction and …
-
Enhancing language models for telecom networks using retrieval-augmented generation: A pipeline for telecom-specific QA system
Natural Language Processing (NLP) is used to address several problems across various domains. There is an increasing demand for domain-specific NLP applications, and telecommunication networks are one of the vital fields of this application. In NLP, generative artificial intelligence (Gen-AI) has …
-
Efficient and scalable multi-objective optimization for foundation models
… that often outperform manually engineered models across diverse tasks. As deep learning systems are deployed at scale, architectural optimality must extend beyond predictive performance to incorporate deployment constraints. This issue is especially pronounced for foundation models, such as …
-
Improving reasoning in (smaller) language models
This thesis investigates how compact language models, despite their limited parameter counts, can achieve robust multi-step reasoning traditionally associated with much larger systems. Instead of pursuing scale, we propose a structured, agentic framework built around three core capabilities: plan, …
Page 1 of 2