2 papers
cs.LG2026
Mechanistic Foundations of Goal-Directed Control
Alma Lago
Mechanistic interpretability has transformed the analysis of transformer circuits by decomposing model behavior into competing algorithms, identifying phase transitions during trai…
cs.AI2024
Object-centric proto-symbolic behavioural reasoning from pixels
Ruben van Bergen, Justus Hübotter, Alma Lago +1
Autonomous intelligent agents must bridge computational challenges at disparate levels of abstraction, from the low-level spaces of sensory input and motor commands to the high-lev…