Source record

Published by arXiv cs.MA (T1 source). The original is at https://arxiv.org/abs/2610.08793.

Pipeline notes

The summary and note below are generated by the signal pipeline — they are Beyond Desk’s reading, not quotations from the source.

Proposes Adaptive Workflow Intelligence (AWI), a four-layer Perception-Cognition-Action-Reflection cognitive architecture for enterprise AI agents. It uses reflective memory and feedback-driven adaptation to handle policy constraints and environmental drift, evaluated in a simulated drift-and-delay stress test where guardrail-constrained adaptive methods recover faster than static automation.

Formalizes a reflection loop for policy-bound enterprise agents and tests it under regime shift, though the reported gains over static baselines are modest and confined to simulation.

Cited by

No citations on record.