4 papers
Surprisal-Guided Selection: Compute-Optimal Test-Time Strategies for Execution-Grounded Code Generation
Jarrod Barnes
Test-time training (TTT) adapts language models through gradient-based updates at inference. But is adaptation the right strategy? We study compute-optimal test-time strategies for…
OpenSec: Measuring Incident Response Agent Calibration Under Adversarial Evidence
Jarrod Barnes
As large language models (LLMs) improve, so do their offensive applications: frontier agents now generate working exploits for under $50 in compute (Heelan, 2026). Defensive incide…
Continual Learning, Not Training: Online Adaptation For Agents
Aman Jaglan, Jarrod Barnes
Continual Learning (CL) methods have traditionally focused on mitigating catastrophic forgetting through gradient-based retraining, an approach ill-suited for deployed agents that…
Reduction of Electromagnetic Interference in ultra-low noise Bimodal MEG & EEG
Jim Barnes, Lukasz Radzinski, Soudabeh Arsalani +4
Single-channel SQUID system technology, operating at a noise level of 100s of aT/, enables the non-invasive detection of synchronized spiking activity at the si…