Context Assembly as the Controlled Variable: A Control-Theoretic View of Harness Policies for Frozen LLM Agents

• URL: http://arxiv.org/abs/2607.25408v1
• Date: Tue, 28 Jul 2026 08:07:06 GMT
• Title: Context Assembly as the Controlled Variable: A Control-Theoretic View of Harness Policies for Frozen LLM Agents
• Abstract summary: We do not claim to introduce control theory to LLM agents -- that ship has sailed.<n>Prior work controls tool selection, inter-agent message routing, or the agent's raw action stream.<n>This paper develops the formal decomposition of the online controller.
• Score: 2.452218069720666
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: A growing body of 2026 work applies control theory to LLM agents: Lyapunov-certified stability for tool-mediated controllers (Prinos et al., "Stable Agentic Control", 2026), sample-complexity bounds for sparse policies over massive discrete tool universes (Majumdar, "Sparse Agentic Control", 2026), and regulatory-control decompositions of multi-agent systems into auditable feedback loops (Nogueira and Skogestad, 2026). We do not claim to introduce control theory to LLM agents -- that ship has sailed. Our narrower claim is about what the controlled variable is. Prior work controls tool selection, inter-agent message routing, or the agent's raw action stream. We instead treat context assembly itself -- which prompt template, which few-shot demonstrations, how much retrieved context, how many planning/verification passes -- as the controlled variable, learned online by a contextual bandit or REINFORCE policy sitting outside a frozen model. This paper develops the formal decomposition (inner frozen policy $π_θ$, outer context policy $π_φ$), gives a stability argument for the online controller in the sense used by Zhang et al. (2026) (non-decreasing expected reward under bounded policy change), and reports an uncertainty-calibration analysis of the controller's own confidence against realized task outcomes. The applied counterpart to this paper instantiates the same controller across three domains and two model providers and releases the dataset, trajectory logs, and a deployment recipe; here we focus on the formal framing and the stability/uncertainty evidence a control-theoretic claim requires.

Related papers

• From Transient Prompts to Persistent Control: Scientific Poster Generation via Recursive Semantic-Geometric Contracts [67.34566880931716]
  PosterVisor is a control framework that shifts poster generation from transient prompts to persistent control.<n>An Orchestrator grounds rubrics in the paper and visual assets, compiling them into a Semantic-Geometric Contract.<n>Recursive Contract Enforcement (RCE) dynamically triggers checks across stages as evidence emerges.
  arXiv  Detail & Related papers  (2026-09-15T15:32:07Z)
• Agentic Visual Generation: From Generative Models to Agentic Control [73.3572965008302]
  In most existing systems, the controller is an LLM or VLM, while visual generation models serve as tools or executors.<n>Planning depth, tool use, multi-role collaboration, and reinforcement learning are often treated as evidence of agenticity.<n>We organize the field according to what the controller can directly control in the generation process.
  arXiv  Detail & Related papers  (2026-09-06T17:59:24Z)
• A Systematic Approach to Multi-Agent AI from Advanced Regulatory Control Theory: Safe and Auditable LLM Operator Agents for Process Control [0.2864713389096699]
  It is possible to hypothesise that a multi-agent reformulation under process-control principles offers a route to address those points.<n>The present work proposes such a reformulation, derived from Advanced Regulatory Control (ARC) theory.
  arXiv  Detail & Related papers  (2026-06-29T20:07:47Z)
• A hardware-safety-gated system for LLM-written native ARTIQ control code on a trapped-ion platform [18.226528651110133]
  Large-language-model (LLM) agents can write and run experimental control code.<n>Unchecked code can damage the apparatus, and there is no formal, per-operation boundary between human authorization/supervision, and agent decisions.<n>We present a control system that places an LLM agent in the loop of a trapped-ion experiment while enforcing such a boundary.
  arXiv  Detail & Related papers  (2026-06-25T16:19:23Z)
• RACL: Reasoning-Agent Control Layers for Continuous Metaheuristic Learning [0.0]
  This paper introduces RACL, a Reasoning-Agent Control Layer for metaheuristics.<n>The agent does not replace the guardrail and does not modify business constraints.<n>It controls the agent's internal search behavior by observing operational memory, reasoning over past behavior, formulating bounded hypotheses, testing interventions, evaluating outcomes, applying algorithmics, consolidating useful policies and explaining its decisions.
  arXiv  Detail & Related papers  (2026-06-18T12:05:16Z)
• Machine-Coached Policy Revision in Adaptive Agent-Based Regulatory Simulation: A Controller-Level Contestability Layer [0.0]
  This paper proposes a lightweight machine-coached policy-revision layer for adaptive agent-based regulation.<n>The contribution is not a new optimal controller but does not claim formal guarantees for unrestricted machine coaching.
  arXiv  Detail & Related papers  (2026-06-15T10:36:10Z)
• Structural Distinguishability of Static and Adaptive Policy Regimes in Agent-Based Regulatory Simulation [0.0]
  This paper contributes a controlled simulation benchmark rather than a new general framework.<n>We evaluate naive fixed policies, tracking-aware fixed policies, and three adaptive controllers: setpoint, safety-margin, and one-sided control.<n>The contribution is methodological: scalar indicators, cap-relative symbolic diagnostics, trajectory motifs, and visual inspection jointly reveal how regulatory conclusions can differ even when average outcomes appear similar.
  arXiv  Detail & Related papers  (2026-06-15T10:06:32Z)
• CIAware-Bench: Benchmarking Control Intervention Awareness Across Frontier LLMs [100.38986535324284]
  We introduce textbfCIAware-Bench, a benchmark for measuring textbfcontrol textbfintervention (CI) awareness across frontier models.<n>CIAware-Bench tests whether models can distinguish their own trajectories from those modified by a control intervention.
  arXiv  Detail & Related papers  (2026-06-09T16:24:16Z)
• Taming Actor-Observer Asymmetry in Agents via Dialectical Alignment [59.536125286960186]
  Multi-agent frameworks assigning specialized roles are increasingly adopted to enable self-reflection and mutual auditing.<n>We find it simultaneously induces a human-like cognitive bias known as Actor-Observer Asymmetry.<n>ReTAS guides agents to synthesize conflicting viewpoints into an objective consensus.
  arXiv  Detail & Related papers  (2026-04-21T15:05:58Z)
• PolicyBank: Evolving Policy Understanding for LLM Agents [51.86716874651299]
  PolicyBank maintains structured, tool-level policy insights and iteratively refines them.<n>PolicyBank closes up to 82% of the gap toward a human oracle.
  arXiv  Detail & Related papers  (2026-04-16T20:29:30Z)
• Dynamic analysis enhances issue resolution [53.50448142467294]
  DAIRA (Dynamic Analysis-enhanced Issue Resolution Agent) is an automated repair framework that embeds dynamic analysis into the agent's reasoning cycle.<n> Driven by a Test Tracing-Driven methodology, DAIRA utilizes lightweight monitors to extract critical runtime data.<n>When powered by Gemini 3 Flash Preview, DAIRA establishes a new state-of-the-art (SOTA) performance, achieving a 79.4% resolution rate on the SWE-bench Verified dataset.
  arXiv  Detail & Related papers  (2026-03-23T14:48:54Z)
• Policy Search for Model Predictive Control with Application to Agile Drone Flight [56.24908013905407]
  We propose a policy-search-for-model-predictive-control framework for MPC. Specifically, we formulate the MPC as a parameterized controller, where the hard-to-optimize decision variables are represented as high-level policies. Experiments show that our controller achieves robust and real-time control performance in both simulation and the real world.
  arXiv  Detail & Related papers  (2021-12-07T17:39:24Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.