https://etd.auburn.edu/handle/10415/2 2026-08-07T22:20:11Z 2026-08-07T22:20:11Z Nahid, Raisul Arefin https://etd.auburn.edu/handle/10415/10635 2026-08-07T21:02:07Z 2026-08-07T00:00:00Z

Type Inference in Stripped Binaries via NLP-Guided Binary Analysis Nahid, Raisul Arefin Binary program analysis poses fundamental challenges due to the loss of high-level abstractions during the compilation process. In contrast to source code, binary executables lack explicit representations of variables, types, control structures, and other abstractions essential for program comprehension. The primary objective of this dissertation is to advance the state of type inference in binaries, a core problem in binary reverse engineering. Accurately reconstructing data type information enables more reliable recovery of source-level representations. At the heart of our approach is the use of Natural Language Processing (NLP) models for type inference. We formulate type prediction as a sequence classification task over instruction-level representations. However, standard NLP architectures and tokenization strategies are not well-suited to the properties of assembly code. We therefore develop and evaluate specialized tokenization methods for assembly, designed to preserve semantic granularity and structural regularity in the input sequences. These methods yield measurable improvements in downstream model performance on type prediction tasks. A further question concerns which class of models is best suited to type inference itself. Type inference can be performed using a smaller transformer trained specifically for the task or a large, general-purpose large language model (LLM) applied without task-specific supervision. To resolve this trade-off, we conduct a systematic evaluation that compares both paradigms across a spectrum of binary analysis tasks centered on type inference. This study clarifies when model scale can substitute for task-specific supervision and provides concrete guidance for selecting the appropriate modeling approach. As the primary direction for future work, we outline an NLP-driven approach to reconstructing high-level type abstractions from C++ binaries. C++ binaries are especially difficult to analyze because features such as classes, templates, and multiple inheritance leave only fragmentary traces in raw binary form. We aim to map low-level assembly instructions to high-level abstractions, including Standard Template Library types (e.g., map, list) and their corresponding class and structure definitions. Recovering these object-oriented abstractions would improve decompilation, vulnerability detection, and the analysis of complex C++ programs. Together, these components—specialized tokenization, transformer-based type inference, and a systematic evaluation of model paradigms—constitute a unified NLP-driven framework for binary program analysis, with improved disassembly reliability identified as a direction for future work.

2026-08-07T00:00:00Z Nahid, Raisul Arefin https://etd.auburn.edu/handle/10415/10634 2026-08-07T21:01:29Z 2026-08-07T00:00:00Z

Type Inference in Stripped Binaries via NLP-Guided Binary Analysis Nahid, Raisul Arefin Binary program analysis poses fundamental challenges due to the loss of high-level abstractions during the compilation process. In contrast to source code, binary executables lack explicit representations of variables, types, control structures, and other abstractions essential for program comprehension. The primary objective of this dissertation is to advance the state of type inference in binaries, a core problem in binary reverse engineering. Accurately reconstructing data type information enables more reliable recovery of source-level representations. At the heart of our approach is the use of Natural Language Processing (NLP) models for type inference. We formulate type prediction as a sequence classification task over instruction-level representations. However, standard NLP architectures and tokenization strategies are not well-suited to the properties of assembly code. We therefore develop and evaluate specialized tokenization methods for assembly, designed to preserve semantic granularity and structural regularity in the input sequences. These methods yield measurable improvements in downstream model performance on type prediction tasks. A further question concerns which class of models is best suited to type inference itself. Type inference can be performed using a smaller transformer trained specifically for the task or a large, general-purpose large language model (LLM) applied without task-specific supervision. To resolve this trade-off, we conduct a systematic evaluation that compares both paradigms across a spectrum of binary analysis tasks centered on type inference. This study clarifies when model scale can substitute for task-specific supervision and provides concrete guidance for selecting the appropriate modeling approach. As the primary direction for future work, we outline an NLP-driven approach to reconstructing high-level type abstractions from C++ binaries. C++ binaries are especially difficult to analyze because features such as classes, templates, and multiple inheritance leave only fragmentary traces in raw binary form. We aim to map low-level assembly instructions to high-level abstractions, including Standard Template Library types (e.g., map, list) and their corresponding class and structure definitions. Recovering these object-oriented abstractions would improve decompilation, vulnerability detection, and the analysis of complex C++ programs. Together, these components—specialized tokenization, transformer-based type inference, and a systematic evaluation of model paradigms—constitute a unified NLP-driven framework for binary program analysis, with improved disassembly reliability identified as a direction for future work.

2026-08-07T00:00:00Z BEKELE, FEVEN https://etd.auburn.edu/handle/10415/10633 2026-08-07T18:51:24Z 2026-08-07T00:00:00Z

Application of natural feed additives to enhance growth performance, innate immunity, gut microbiome, and disease resistance of Nile tilapia (Oreochromis niloticus) raised in biofloc and recirculating clear water systems BEKELE, FEVEN With global demand for animal protein growing, aquaculture has shifted toward more intensive production. However, intensified systems are often challenged by disease outbreaks and water quality issues. Biofloc technology improves water quality through microbial activity and can reduce feed costs and wastewater discharge, while functional feed additives offer promising alternatives to antibiotics. The first part of this project evaluated the effects of dietary supplementation with humic acid (HA) and fulvic acid (FA) in Nile tilapia (Oreochromis niloticus) reared in a biofloc system. Fish fed HA showed improved growth performance, and both HA and FA increased gut microbial diversity and altered community structure. A second trial assessed graded inclusion levels of a commercial phytogenic additive in tilapia challenged with Streptococcus agalactiae and Flavobacterium oreochromis. Growth performance was unaffected, but the highest inclusion level improved survival following F. oreochromis challenge. These findings support the use of functional feed additives as practical alternatives to antibiotics in tilapia culture.

2026-08-07T00:00:00Z Gosier, Glen Sr https://etd.auburn.edu/handle/10415/10632 2026-08-07T18:27:14Z 2026-08-07T00:00:00Z

The Merger of the Future Farmers of America and the New Farmers of America in 1965: A Phenomenological Case Study Gosier, Glen Sr This qualitative single-case study examined a purposive sample of six participants, focusing on how their experiences as members of NFA during the period in question were retrospectively interpreted. The participation of African Americans in agricultural education has evolved through significant historical and institutional change. The Future Farmers of America, founded in 1928, preceded the establishment of the New Farmers of America in 1935, which was created to support African American male students through leadership preparation and agricultural instruction in segregated settings. In response to desegregation efforts following the Civil Rights Act of 1964, the NFA and FFA merged in 1965, thereby ending the NFA as a separate entity and establishing a single integrated organizational structure in agricultural education. Using phenomenologically oriented interviews and reflexive thematic analysis, this study explored how participants interpreted the legacy of that merger. The study was guided by Structural Discrimination and discussion of identity and belonging within a constructivist paradigm, which assumes that meaning is socially constructed through lived experience and interpretation. Findings were organized into five themes: perceived beneficial effects of the 1965 merger on agricultural education, perceived negative effects of the merger, loss of African American identity, experiences and perceptions of racial discrimination, and the future of agricultural education for African Americans. The findings suggest that participants viewed the merger as a historically complex and uneven process. On one hand, they associated integration with expanded access to scholarships, contests, leadership development, and professional opportunities. On the other hand, they connected the merger to the erosion of Black educational spaces, the decline of African American agriculture teachers, the loss of NFA traditions, and the persistence of racial inequities within agricultural education. Participants’ accounts therefore indicate that formal inclusion did not necessarily result in equality, belonging, or shared institutional power. This study contributes to the literature by centering participant interpretations of how integration reshaped African American experiences in agricultural education. In doing so, it highlights the importance of representation, mentorship, historical recognition, and institutional accountability in efforts to strengthen the future participation of African Americans in the profession. It also offers theoretical, practical, and research implications for understanding the continuing legacy of the NFA-FFA merger.

2026-08-07T00:00:00Z