SOURCE-LINKED INTELLIGENCE

Polished but Unresolved: Identifying Late-Stage Pressure States in Long-Horizon Tool-Use Agents

arXiv · AI, language, vision and robotics · article · Sep 1, 2026 · UTC

Long-horizon tool-use agents need not only to search and plan, but also to decide when to finalize. We study late-stage pressure states, in which an agent is biased toward submitting a final answer that appears complete and polished while key constraints remain unresolved. We first train a linear probe to show that this pressure state is identifiable from the agent's hidden states. Then, we use activation interventions along this pressure direction and find that shifting the hidden states changes both the pressure score and whether the agent continues tool use or submits early. Through control

Read original source ↗ Open in workspace

recordType

paper

region

Global