{"ID":2833389,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2512.03571","arxiv_id":"2512.03571","title":"EnCompass: Enhancing Agent Programming with Search Over Program Execution Paths","abstract":"We introduce a new approach to agent programming, the development of LLM-based agents. Current approaches to agent programming often entangle two aspects of agent design: the core workflow logic and the inference-time strategy (e.g., tree search). We introduce \"probabilistic angelic nondeterminism\" (\"PAN\"), a programming model that disentangles these two concerns, allowing the programmer to describe the agent workflow and independently experiment with different inference-time strategies by simply changing a few inputs. We provide an implementation of PAN in Python as the EnCompass framework, which uses a Python decorator to compile agent workflow programs into a search space. We present three case studies that demonstrate how the framework lets the programmer quickly improve the reliability of an agent and easily switch between different inference-time strategies, all with little additional coding.","short_abstract":"We introduce a new approach to agent programming, the development of LLM-based agents. Current approaches to agent programming often entangle two aspects of agent design: the core workflow logic and the inference-time strategy (e.g., tree search). We introduce \"probabilistic angelic nondeterminism\" (\"PAN\"), a programmi...","url_abs":"https://arxiv.org/abs/2512.03571","url_pdf":"https://arxiv.org/pdf/2512.03571v1","authors":"[\"Zhening Li\",\"Armando Solar-Lezama\",\"Yisong Yue\",\"Stephan Zheng\"]","published":"2025-12-03T08:50:16Z","proceeding":"cs.AI","tasks":"[\"cs.AI\",\"cs.LG\",\"cs.PL\"]","methods":"[\"Large Language Model\"]","has_code":false}