# The Context Gathering Decision Process: A POMDP Framework for Agentic Search Source: [The Context Gathering Decision Process: A POMDP Framework for Agentic Search]() Collection index: [llms.txt]() Data updated: 2026-05-12T05:00:04.827Z Authors: Chinmaya Kausik, Adith Swaminathan, Nathan Kallus Published: 2026-05-07T23:45:07.000Z ## Abstract Large Language Model (LLM) agents are deployed in complex environments -- such as massive codebases, enterprise databases, and conversational histories -- where the relevant state far exceeds their context windows. To navigate these spaces, an agent must iteratively explore the environment to find relevant information. However, without explicit infrastructure, an agent's working memory can degrade into lossy representations of the search state, resulting in redundant work (e.g. repetitive looping) and premature stopping. In this work, we formalize this challenge as the Context Gathering Decision Process (CGDP), a specialized Partially Observable Markov Decision Process, where an agent's objective is to adaptively refine its belief state to isolate the necessary information for a task. We model an LLM's behavior as approximate Thompson Sampling within this CGDP, and introduce a predicate-based method that decomposes an LLM's implicit search into explicit and modular operations. We then derive two plug-and-play interventions for iterative LLM agents: a persistent, predicate-based belief state that bounds context while preserving multi-hop reasoning, and a programmatic exhaustion gate that halts unproductive search without premature stopping. Across four methods and three question-answering domains, we empirically validate that replacing an LLM's implicit state with our CGDP-motivated belief state improves multi-hop reasoning by up to $11.4\\%$; while the modular programmatic exhaustion detection saves up to $39\\%$ of tokens without any degradation in agent performance. Ultimately, we argue that framing the LLM agent loop as a CGDP can guide the design of modular, non-interfering improvements to agentic search harnesses. ## Recorded benchmark results A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage. - [Multi-hop Question Answering on MuSiQue (Rubric Judge Accuracy)](): Multi-hop Question Answering; MuSiQue; Rubric Judge Accuracy: 39.6. - [Software Engineering Question Answering on SWE-QA Pro](): Software Engineering Question Answering; SWE-QA Pro; Rubric Judge Accuracy: 51.4. - [Long-context Question Answering on LoCoMo (Rubric Judge)](): Long-context Question Answering; Locomo; Rubric Judge Accuracy: 66.6. ## Original sources and implementation [Original paper]() [PDF]()