STAGE, an executable-graph framework that confines model judgment to policy-scoped nodes while placing procedural control in deterministic code, improves task success and repeated-run reliability compared with monolithic full-policy execution.
Abstract
Policy-governed agents must interpret case evidence while reliably following authorized procedures. We present STAGE, an executable-graph framework that confines model judgment to policy-scoped nodes while placing procedural control in deterministic code. We evaluate STAGE on three public policy-following benchmarks and Smart Dispute, a proprietary banking benchmark. Compared with monolithic full-policy execution, STAGE improves task success and repeated-run reliability, with its largest observed gains on the deeper workflows. On $\tau^2$-bench Telecom and Smart Dispute, $\mathrm{Pass}^{3}$ improves by up to 55.0 and 65.7 percentage points, respectively. These results demonstrate the value of combining localized policy reasoning with deterministic procedural control for enterprise use.
We study a governed approach to enterprise analytics: a language model interprets the question, while deterministic policy selects and runs a pre-approved analytical program that returns both results and evidence. We show that this restriction can remain expressive within a defined analytical class, using relational op...
This work identifies a state-transmission failure between information extraction and action in large language model agents, and shows how handoff transformations can retain state content while weakening its constraints on downstream action.
Yi-Heng Sun, Hui-Fei Wang, Yan-Cheng Zhu et al.· 2 citations
PolicyGuide compiles each domain policy into a workflow graph and invokes a proactive verifier at user-turn boundaries and finds the lowest observed attack-success rate under adversarial users and the strongest procedural compliance in an author-designed workflow-level validation.
Seongjae Kang, Taehyung Yu, Sung Ju Hwang· 2 citations
This work proposes 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.
Eluna is a graph-guided, multi-agent framework that encodes SOPs as directed acyclic graphs with progressive disclosure and delegates independent tasks to parallel sub-agents, each with persistent code execution and live data access.
Ning Liu, Kalle Kujanpää, Zhaoxuan Zhu et al.· 1 citation
COVENANT is proposed, a compiler-and-interpreter architecture for workflow-aligned agent execution that substantially mitigates workflow misalignment, moving LLM-agent alignment beyond isolated prompt following toward reliable execution of complex and multi-step workflows.
Jin-Cheng Wang, Ming-Kai Zheng, Tao Wei· arXiv.org· 1 citation
Related blog posts
MIT News · Artificial Intelligence· news.mit.eduOct 13, 2026
Machine-learning networks can learn without detailed instructions from a designer. Those that do may offer insight into how the brain reorganizes itself during learning.
With $2.1 million funding from Google.org, the open-source Public Transit Intelligence Hub will unify public transit monitoring, operations, and passenger communication.
