Skip to main content

2023 Research Symposium

Each year, LAS undertakes a research program involving partners from a variety of academic, industry, and government communities. The outcomes of these research projects are of interest to our intelligence community stakeholders, as well as the respective communities of our academic and industry partners.


We invite you to learn about this year’s unclassified research projects. Projects are grouped into three themes: Human-Machine Teaming, Operationalizing AI/ML and Content Triage. Descriptions for each project will be added in December.

Human-Machine Teaming

This theme encompasses efforts toward enhancing the effectiveness of human analysts partnering with emerging automated technologies. We consider both the human user’s experience with teaming technologies and the creation of technologies to alleviate pain points and reduce cognitive load for analysts. Though technological innovations have the potential to reduce the cognitive burden associated with processing and managing vast datasets, their successful implementation hinges on a deep understanding of how humans can process the outputs and seamlessly integrate these technologies into their workflows. Machine teammates have significant potential to enhance efficiency on tasks; however, they may not be effective unless grounded in a robust model of how human users carry out tasks and the cognitive needs of the user.

Within this theme, we group projects by the following topic areas:

Modeling Users & Tradecraft

These projects examine what a human user needs and does when making decisions in different tasks and settings.

Calibrating Trust

These projects explore the interactions and interdependencies between human and machine, enabling the human to evaluate information in context.

Generating & Exploring Hypotheses

These analytic workflows integrate machine capabilities to synthesize information for different tasks.

Promoting Cognitive Engagement

These projects leverage AI capabilities to enhance the ability of the human user to interact with complex data.

Operationalizing AI & ML

LAS research on machine learning (ML) and artificial intelligence (AI) focuses on how machine learning concepts and techniques can still be useful even when working under the constraints of an operational environment. Researchers examine the impact of those constraints on AI/ML performance when applied in operational settings. They also seek methods to mitigate the costs of those constraints, whether they be financial, time, or cognitive resources.

These efforts include creating data from which models can learn, training models, running models in realistic conditions, and sharing models for widespread use.

Operational constraints might include limited amounts of quality data available or competing priorities for subject matter experts who can annotate data for domain-specific tasks; limited computational resources for training models; highly variable conditions in which AI/ML might be deployed; and logistical challenges around running models outside of the team or environment in which they were developed.

Within this theme, we group projects by the following topic areas:

Amplifying Knowledge

Increasing Efficiency

Expanding Usability

Overcoming Variability

Content Triage

This theme has evolved to focus on mission needs for image, audio, and text, while still addressing the historical Content Triage driver to focus on large volumes of data. Content Triage is also pushing the boundaries of machine learning technology by tackling multimodal challenges that track challenging modern communications systems and data. In 2023, Content Triage research examined new, efficient methods to access and exploit large amounts of data. These new methods include semantic queries that help an analyst better navigate unknown information, improved use of knowledge graph structures, better ways to characterize voice, and flexible detection of sounds and images – all to enable analysts to retrieve unknown insights from data. After all, useful information that systems contain but are never aware of is perhaps the most tragic missed opportunity for the intelligence community’s prodigious efforts.

Content Triage projects are divided into three scopes to explore at our symposium: Sight, Sound, and Search. These innovative projects demonstrate novel ways to address mission challenges around the ever-present need to process and exploit large data volumes.