Absorbed in Inertia: Activation Analysis for Computer-Use Agents

The paper investigates a phenomenon called inertia in computer‑use agents, where agents repeat ineffective actions despite recognizing their futility. By analyzing high‑dimensional activation states, the authors find that inertia corresponds to an absorbing region in activation space where values become stale. They propose a method called R$^3$—Reset, Reroute, Restore—to temporarily reset the agent’s context trajectory, escape the absorbing region, and then restore historical context, achieving a 17‑55% reduction in measured inertia across models.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv AI.

• agents
• multimodal