AMBER: Training Long-Horizon Web Agents through Append-Only Memory

Excerpt

arXiv:2610.07118v1 Announce Type: cross Abstract: Modern language-model agents increasingly interact with external environments over long-horizon, multi-step trajectories, where the accumulated interaction history can quickly exceed practical context budgets. To ensure reliability, agents must maintain factual information over long horizons, remember execution errors and corrective feedback, and track progress across actions. Several approaches have been proposed to achieve this without the need