Brandon
Colelough



Ph.D. candidate | AU Fulbright Scholar

ADF Officer | Electrical Engineer

Biography

Brandon Colelough is a PhD candidate in Computer Science at the University of Maryland and a 2023 Australian Fulbright Scholar, researching neuro-symbolic AI for explainable and trustworthy systems under Professor William Regli. Brandon Colelough also serves as a research fellow at the NIH's National Library of Medicine, where his work on hallucination detection and mitigation in large language models targets the same challenge that drives his doctoral research, enabling trusted autonomy in high-risk environments. Brandon Colelough is also a Signals Corps officer in the Australian Army and a chartered electrical engineer.

Education

University of Maryland

College Park

Ph.D. candidate

2023 - Present

ADF School of Signals

Melboure

ADF Signals Corps Officer

2023

University of New South Wales

Canberra

Electrical Engineering (Hons)

2018-2022

Royal Military College - Duntroon

Canberra

Australian Army GSO

2021

Australian Defence Force Academy

Canberra

Professional Military Education

2018 – 2020

School of Technical Training

Wagga Wagga, Sydney

Aerospace Technical Trade

2017

Employment

National Institutes of Health

Postgraduate Researcher

2024-Present

UMD Department of CS

Teaching Assistant and Graduate Researcher

2023-2024

Australian Defence Force

ADF Signals Corps Officer

2017-Present

Power Earth Engineering

Electrical Engineer

2022

UNSW SEIT

Research Assistant

2019 - 2022

Royal Australian Air Force

Aircraft Technician

2017

Publications

Neuro-Symbolic Artificial Intelligence: A Reference Model for the Design and Development of Neuro-Symbolic Systems

A reference model for the design and development of Neuro-Symbolic systems

Author: Brandon C. Colelough, William Regli

Coming soon Paper

Quantifying Hallucinations in Large Language Models on Medical Textbooks

An experimental study enabled by ClinIQLink

Author: Brandon C. Colelough, Davis Bartels, Dina Demner-Fushman

arXiv Paper

Overview of the ClinIQLink 2025 Shared Task on Medical Question-Answering

BioNLP 2025 shared task overview for medical question-answering

Author: Brandon C. Colelough, Davis Bartels, Dina Demner-Fushman

arXiv ACL Anthology Paper

Towards Grounded Hallucination Definitions for Biomedical Question Answering with Reproducible Examples from ClinIQLink

A layered definition framework for biomedical QA hallucinations

Author: Brandon C. Colelough, Davis Bartels, Madeline Bittner, Dina Demner-Fushman

Coming soon Paper

Relations of Linguistic Features to Guideline-Based Rankings of Medical Answers are Nontrivial

An exploratory analysis of linguistic features in medical answer evaluation

Author: Davis Bartels, Brandon C. Colelough, Dina Demner-Fushman

Coming soon Paper

Beyond the Prompt: Approaches to Knowledge Boundary Detection in Large Language Models - A Systematic Literature Review

A systematic review of introspective knowledge boundary detection in LLMs

Author: Brandon Colelough, Davis Bartels, Dina Demner-Fushman, Ishan Tamrakar, Kwesi Cobbina, Mike Ledford, Srividya Ponnada, Xinchen Yang, Yuexi Chen

Coming soon Paper

Reproducibility in Computer Science: An Empirical Multi-Year Replication Study of Neuro-Symbolic AI

An empirical reproducibility study of the Neuro-Symbolic AI literature

Author: Brandon Colelough, Vladimir Martirosyan, Ishan Tamrakar, William Regli, Aditya Kumar, Anh N. Nhu, Dhruv Dubey, Raj Ambavane, Haowei Deng

Coming soon Paper

LNSAI 2024 - Survey Paper

Neuro-Symbolic AI in 2024: A systematic review

Author: Brandon Colelough, William Regli

arXiv Paper

ClearComm

Effects of Dataset Sampling Rate for Noise Cancellation through Deep Learning

Author: Brandon Colelough

arXiv Paper

SymboSLAM

Semantic Map Generation in a Multi-Agent System

Author: Brandon Colelough

arXiv Paper

Advancing Frontiers in SLAM

A Survey of Symbolic Representation and Human-Machine Teaming in Environmental Mapping

Author: Brandon Colelough

arXiv Paper

AEMO Model

Australian Energy Market Operator National Electricity Market Network Optimal Power Flow Modelling

Author: Brandon Colelough

arXiv Paper

Enhanced Thermal Management in High-Temperature Applications

Design and Optimization of a Water-Cooled Forced Convection System in a Hollow Cuboid Vapour Chamber Using COMSOL and MATLAB

Author: Brandon Colelough

arXiv Paper

BlackBoard

Advancing Autonomous Target Detection and Mapping: Development of Integrated Software Systems for Monocular Depth Sensing and Dynamic Target Labeling

Author: Brandon Colelough

arXiv Paper

Sheep Labelling Image Dataset

Optimizing Data Diversity for Accurate Prediction Models: Insights from a Multi-Class Sheep Image Data-Set Analysis towards Autonomous Shepherding

Author: Brandon Colelough

arXiv Paper

Projects

ClinIQLink

A medical question-answering benchmark and shared task for evaluating medically oriented language-model systems with expert-verified, source-grounded questions.

Website BioNLP GitHub

MechaTerp

MechaTerp is a comprehensive tool for mechanistic interpretability analysis of large language models, specifically designed to detect semantic path divergences and analyze model behavior deviations.

Project GitHub

Karaoke Machine

An AI-powered karaoke generator that separates vocals from music, generates time-synced lyrics, and displays them in real time during accompaniment playback.

Project

Dance Machine

A split Studio and Play platform for dance ingestion, review, publishing, learner practice, camera capture, feedback, scoring, and future game-mode experiences.

Project

RAG-DAT

A closed-corpus retrieval-augmented question-answering system that grounds answers in verbatim quotes with traceable document, page, section, span, and bounding-box provenance.

Project

LIO

An open hardware experiment to build a low-cost, 3D-printable multispectral diffuse-reflectance scanner for powdered food samples and controlled mixture experiments.

Project page coming soon

In the News

Interviewed & quoted

How good old-fashioned AI could spark the field's next revolution

Nature — News Feature

Read Article
Interviewed

Video Interview

YouTube

Watch Video
Quoted

Researcher Warns That AI Hallucinations Are Unavoidable

The Hoya — Georgetown University

Read Article
Featured

Australia, United States team up on AI

Australian Department of Defence

Read Article
Featured

UNSW Canberra alumnus Lieutenant Brandon Colelough is paving the way for artificial intelligence research as 2023 Fulbright Scholar

UNSW Canberra

Read Article
Profile

Brandon Colelough — UMD Department of Computer Science

University of Maryland

View Profile
Email Icon Email Resume Icon Resume LinkedIn Icon LinkedIn GitHub Icon GitHub Instagram Icon Instagram