Hierarchical Prompt-Domain Control and Learning for Resource-Constrained Agentic Language Models

• URL: http://arxiv.org/abs/2605.27703v1
• Date: Tue, 26 May 2026 21:23:30 GMT
• Title: Hierarchical Prompt-Domain Control and Learning for Resource-Constrained Agentic Language Models
• Abstract summary: Large Language Models are increasingly deployed inside agentic systems.<n>We propose a hierarchical control-and-learning framework in which a compact model is first distilled to learn the required output schema.<n>We show improved reliability and cost-efficiency over non-hierarchical, distillation-only, and non-distilled baselines.
• Score: 0.0
• License: http://creativecommons.org/licenses/by-nc-nd/4.0/
• Abstract: Large Language Models are increasingly deployed inside agentic systems, where they must follow structured protocols, adapt to evolving states, and operate under memory, latency, and cost constraints. In such regimes, prompt extension is unreliable: growing contexts can push compact models outside their effective prompt domain, while deployment-time fine-tuning remains limited by scarce data and compute. We propose a hierarchical control-and-learning framework in which a compact model is first distilled to learn the required output schema, then supervised online by an oracle-controller loop. The controller monitors protocol validity and semantic performance, projects accumulated histories into a feasible prompt domain, and triggers lightweight oracle-supervised fine-tuning under drift. This separates schema learning for communication compatibility from semantic adaptation for task-level correction. We formalize prompt-domain feasibility and attention-induced saturation, motivating control of the effective prompt state rather than reliance on nominal context length. Using Multi-Fidelity Bayesian Optimization as a controlled sequential testbed, we characterize a core deployment failure mode and show improved reliability and cost-efficiency over non-hierarchical, distillation-only, and non-distilled baselines.

Related papers

• A Systematic Post-Train Framework for Video Generation [76.26555417456773]
  Large-scale video diffusion models have demonstrated impressive capabilities in generating high-resolution and semantically rich content.<n>A significant gap remains between their pretraining performance and real-world deployment requirements due to critical issues such as prompt sensitivity, temporal inconsistency, and prohibitive inference costs.<n>We propose a comprehensive post-training framework that systematically aligns pretrained models with user intentions through four synergistic stages.
  arXiv  Detail & Related papers  (2026-04-28T09:34:51Z)
• CogniConsole: Externalizing Inference-Time Control as a Formal Abstraction for Reliable LLM Interactions [3.8673630752805437]
  Inference-time control is the computational layer governing task framing and context selection.<n>We introduce emphCogniConsole, an architectural instantiation that externalizes this control into a structured interface.<n>Our results indicate that many observed failure modes, such as context drift and inconsistent constraint adherence, arise from under-specified control rather than insufficient capability.
  arXiv  Detail & Related papers  (2026-04-21T17:21:06Z)
• Gated Condition Injection without Multimodal Attention: Towards Controllable Linear-Attention Transformers [69.60693631859411]
  This paper explores controllable diffusion models built upon linear attention architectures.<n>We propose a novel controllable diffusion framework tailored for linear attention backbones like SANA.<n>Our approach achieves state-of-the-art controllable generation performance based on linear-attention models.
  arXiv  Detail & Related papers  (2026-03-29T12:31:34Z)
• RuntimeSlicer: Towards Generalizable Unified Runtime State Representation for Failure Management [23.96013849484881]
  Metrics, traces, and logs provide complementary views of system runtime behavior.<n>Slicer pre-trains a task-agnostic representation model that encodes metrics, traces, and logs into a single, aligned system-state embedding.<n>State-Aware Task-Oriented Tuning performs unsupervised partitioning of runtime states and enables state-conditioned adaptation for downstream tasks.
  arXiv  Detail & Related papers  (2026-03-23T02:35:13Z)
• Embodied Foundation Models at the Edge: A Survey of Deployment Constraints and Mitigation Strategies [3.4385853021313593]
  Deployment Gauntlet organizes constraints into eight coupled barriers.<n> Reliable deployment depends on system-level co-design across memory, scheduling, communication, and model architecture.
  arXiv  Detail & Related papers  (2026-03-16T21:27:03Z)
• Adaptive RAN Slicing Control via Reward-Free Self-Finetuning Agents [26.34195733862222]
  We propose a novel self-finetuning framework that enables agentic systems to learn continuously through direct interaction with the environment.<n>Our framework implements a bi-perspective reflection mechanism that generates autonomous linguistic feedback to construct preference datasets from interaction history.<n>A subsequent preference-based fine-tuning process distills long-horizon experiences into the model's parameters.<n> Experimental results show that our framework outperforms standard Reinforcement Learning (RL) baselines and existing Large Language Model (LLM)-based agents in sample efficiency, stability, and multi-metric optimization.
  arXiv  Detail & Related papers  (2026-03-11T09:14:56Z)
• ScaleWeaver: Weaving Efficient Controllable T2I Generation with Multi-Scale Reference Attention [86.93601565563954]
  ScaleWeaver is a framework designed to achieve high-fidelity, controllable generation upon advanced visual autoregressive( VAR) models.<n>The proposed Reference Attention module discards the unnecessary attention from image$rightarrow$condition, reducing computational cost.<n>Experiments show that ScaleWeaver delivers high-quality generation and precise control while attaining superior efficiency over diffusion-based methods.
  arXiv  Detail & Related papers  (2025-10-16T17:00:59Z)
• The Gatekeeper Knows Enough [0.0]
  Gatekeeper Protocol is a domain-agnostic framework that governs agent-system interactions.<n>Our results show that this approach significantly increases agent reliability, improves computational efficiency by minimizing token consumption, and enables scalable interaction with complex systems.
  arXiv  Detail & Related papers  (2025-10-16T17:00:42Z)
• Dynamic Generation of Multi-LLM Agents Communication Topologies with Graph Diffusion Models [99.85131798240808]
  We introduce a novel generative framework called textitGuided Topology Diffusion (GTD)<n>Inspired by conditional discrete graph diffusion models, GTD formulates topology synthesis as an iterative construction process.<n>At each step, the generation is steered by a lightweight proxy model that predicts multi-objective rewards.<n>Experiments show that GTD can generate highly task-adaptive, sparse, and efficient communication topologies.
  arXiv  Detail & Related papers  (2025-10-09T05:28:28Z)
• PowerGrow: Feasible Co-Growth of Structures and Dynamics for Power Grid Synthesis [75.14189839277928]
  We present PowerGrow, a co-generative framework that significantly reduces computational overhead while maintaining operational validity.<n> Experiments across benchmark settings show that PowerGrow outperforms prior diffusion models in fidelity and diversity.<n>This demonstrates its ability to generate operationally valid and realistic power grid scenarios.
  arXiv  Detail & Related papers  (2025-08-29T01:47:27Z)
• Lightweight Task-Oriented Semantic Communication Empowered by Large-Scale AI Models [66.57755931421285]
  Large-scale artificial intelligence (LAI) models pose significant challenges for real-time communication scenarios.<n>This paper proposes utilizing knowledge distillation (KD) techniques to extract and condense knowledge from LAI models.<n>We propose a fast distillation method featuring a pre-stored compression mechanism that eliminates the need for repetitive inference.
  arXiv  Detail & Related papers  (2025-06-16T08:42:16Z)

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.