Abstract
Deployed LLM agents rely on agentic context, the model-external textual control content assembled by an operational harness. In this work, the mutable component of that context is a persistent system-level instruction that is updated from operational experience while the model, tools, and harness remain fixed.
Over long evolution horizons, flat-text maintenance makes verification increasingly difficult as accumulated instructions grow and interact.
We propose Graph-Regularized Agentic Context Evolution (GRACE), which maintains the persistent instruction component as a typed semantic graph and validates proposed updates within the local typed neighborhoods of modified nodes. Accepted graph updates are reconstructed as incremental edits to the textual instruction checkpoint used at deployment.
We evaluate GRACE within a fixed telecom agent harness derived from the $\tau^2$-bench under a controlled distribution-shift protocol. Across five independent replications, GRACE improves strict reliability, measured by pass^3, from the Gemini 2.5 Flash zero-shot value of 0.091 to 0.673$\pm0.136$ at the final checkpoint. This exceeds a Gemini 3.1 Pro zero-shot reference of 0.242 on the same held-out set, while the flat-text HCE baseline finishes at 0.191$\pm0.051$. These results identify two requirements for reliable long-horizon context evolution, a structural substrate that makes verification local and a consolidation mechanism that keeps accumulated instruction content usable.
Blogger's Review: The GRACE method effectively addresses the verification challenges in long-term context evolution by structuring instruction content into a semantic graph. Its impressive performance in the telecom domain highlights its robust adaptability in the face of distribution shifts, providing new insights for LLM applications.