Research

The core challenge of intelligence is determining which past experiences are relevant to the current moment. If an agent assumes too much similarity, it over-generalizes. If it assumes too little, it fails to leverage its prior learning.

My work investigates how latent cause inference acts as a gatekeeper for this process. I model how the mind and artificial agents use context, curricula, and task similarity to decide whether to update an existing representation or create a new one. Understanding this decision boundary is crucial for understanding how and when agents fail on meaningful tasks.

Hallucinations in LLMs are a natural consequence of lossy compression, like how humans have false memories. Just as human false memories occur when we reconstruct the past based on "gist" rather than verbatim detail, models hallucinate when their internal representations collapse distinct details into a single, shared prototype.

Autonomous agents can fail in spectacular ways—entering "rumination loops" or decoupling from their environment.

My work proposes that these states have a distinct information-theoretic signature. By monitoring the Entropy Rate of the agent's internal monologue and its Mutual Information with environmental observations, we can detect decoupling in real-time. This offers a path toward reliable "watchdog" systems that can intervene before an agent goes off the rails.

There is a cost in using cognitive control. Allocating it effectively is key to both human skill acquisition and efficient machine learning.

I investigate how agents compute the Expected Value of Control (EVC) to decide when to exact effort versus rely on habit. In AI systems, this translates to dynamic compute allocation—knowing when a "System 2" deep-dive is necessary versus a "System 1" rapid response. This balance is critical for scaling performance without exponential cost.

Scientific discovery often involves navigating high-dimensional search spaces. We are building metascience tools to map these design spaces systematically.

By formalizing experimental design as a search problem, we can use automated methods to identify "blank spots" in the literature—regions of the parameter space that have been theoretically neglected but are physically plausible. This approach accelerates discovery by guiding researchers toward high-value, unobserved phenomena.