Decision-Centric Design for LLM Systems

• URL: http://arxiv.org/abs/2604.00414v1
• Date: Wed, 01 Apr 2026 02:57:23 GMT
• Title: Decision-Centric Design for LLM Systems
• Abstract summary: We propose a decision-centric framework that separates decision-relevant signals from the policy that maps them to actions.<n>This separation supports attribution of failures to signal estimation, decision policy, or execution.<n>Across three experiments, the framework reduces futile actions, improves task success, and reveals interpretable failure modes.
• Score: 3.400316337034188
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: LLM systems must make control decisions in addition to generating outputs: whether to answer, clarify, retrieve, call tools, repair, or escalate. In many current architectures, these decisions remain implicit within generation, entangling assessment and action in a single model call and making failures hard to inspect, constrain, or repair. We propose a decision-centric framework that separates decision-relevant signals from the policy that maps them to actions, turning control into an explicit and inspectable layer of the system. This separation supports attribution of failures to signal estimation, decision policy, or execution, and enables modular improvement of each component. It unifies familiar single-step settings such as routing and adaptive inference, and extends naturally to sequential settings in which actions alter the information available before acting. Across three controlled experiments, the framework reduces futile actions, improves task success, and reveals interpretable failure modes. More broadly, it offers a general architectural principle for building more reliable, controllable, and diagnosable LLM systems.

Related papers

• 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)
• DCFA: Dual-view Causal-inspired Attribution for Failure Reasoning in LLM-based Multi-agent Systems [61.11435469308521]
  We propose DCFA, a training-free framework for failure attribution in large language model systems.<n> DCFA integrates a global module that constructs structured causal-inspired dependency graphs from system traces.<n>Experiments show that DCFA improves step-level accuracy by up to 8.27% over state-of-the-art baselines.
  arXiv  Detail & Related papers  (2026-09-04T05:29:31Z)
• Explanation-Based Runtime Verification for Trustworthy ML-driven Optical Networks [1.0849016186013645]
  We introduce explanation-based runtime verification to assess the soundness of individual machine learning decisions.<n>We show that explanation-based verification can intercept a significant fraction of erroneous decisions while preserving high automation rate.
  arXiv  Detail & Related papers  (2026-07-22T19:22:41Z)
• A Tutorial on Autonomous Fault-Tolerant Control Using Knowledge-Grounded LLM Agents [0.21313236831585222]
  This paper examines how Large Language Model (LLM) agents can support Fault Recovery decisions.<n>The proposed framework treats the LLM as a constrained supervisory planner.<n>It uses plant-specific knowledge to propose recovery actions, and every proposal is checked by an external validator.
  arXiv  Detail & Related papers  (2026-06-30T13:19:45Z)
• OLIVIA: Online Learning via Inference-time Action Adaptation for Decision Making in LLM ReAct Agents [74.20327254615854]
  Large language model agents interleave reasoning, action selection, and observation to solve sequential decision-making tasks.<n>Existing inference-time adaptation methods for LLM agents mainly rely on prompting or retrieval.<n>We propose OLIVIA, an inference-time action adaptation framework for ReAct-style agents.
  arXiv  Detail & Related papers  (2026-05-11T19:28:20Z)
• MultiHedge: Adaptive Coordination via Retrieval-Augmented Control [0.0]
  MultiHedge is a hybrid architecture where an LLM produces structured allocation decisions conditioned on retrieved historical precedents.<n>Our paper contributes a controlled computational study showing that memory and architectural design play a central role in robustness in modular decision systems.
  arXiv  Detail & Related papers  (2026-04-27T18:38:22Z)
• Evaluating Large Language Models for Detecting Architectural Decision Violations [39.480626944948376]
  Architectural Decision Records (ADRs) play a central role in maintaining software architecture quality.<n>Recent advances in Large Language Models (LLMs) open up new possibilities for automating architectural reasoning at scale.<n>We investigated how effectively LLMs can identify decision violations in open-source systems by examining their agreement, accuracy, and inherent limitations.
  arXiv  Detail & Related papers  (2026-02-07T16:36:14Z)
• From "Aha Moments" to Controllable Thinking: Toward Meta-Cognitive Reasoning in Large Reasoning Models via Decoupled Reasoning and Control [11.321315058502215]
  Large Reasoning Models (LRMs) have demonstrated a latent capacity for complex reasoning by spontaneously exhibiting cognitive behaviors such as step-by-step reasoning, reflection, and backtracking, commonly referred to as "Aha Moments"<n>However, such emergent behaviors remain unregulated and uncontrolled, often resulting in overthinking, where the model continues generating redundant reasoning content even after reaching reliable conclusions.<n>Current models are unable to monitor and adaptively manage their reasoning process to determine when to continue, backtrack, or terminate.<n>We propose the Meta-cognitive Reasoning Framework (MERA), which explicitly decouples the thinking process into distinct
  arXiv  Detail & Related papers  (2025-08-06T13:59:17Z)
• Embodied Agent Interface: Benchmarking LLMs for Embodied Decision Making [85.24399869971236]
  We aim to evaluate Large Language Models (LLMs) for embodied decision making.<n>Existing evaluations tend to rely solely on a final success rate.<n>We propose a generalized interface (Embodied Agent Interface) that supports the formalization of various types of tasks.
  arXiv  Detail & Related papers  (2024-10-09T17:59:00Z)
• Cognitive LLMs: Towards Integrating Cognitive Architectures and Large Language Models for Manufacturing Decision-making [51.737762570776006]
  LLM-ACTR is a novel neuro-symbolic architecture that provides human-aligned and versatile decision-making. Our framework extracts and embeds knowledge of ACT-R's internal decision-making process as latent neural representations. Our experiments on novel Design for Manufacturing tasks show both improved task performance as well as improved grounded decision-making capability.
  arXiv  Detail & Related papers  (2024-08-17T11:49:53Z)
• LanguageMPC: Large Language Models as Decision Makers for Autonomous Driving [84.31119464141631]
  This work employs Large Language Models (LLMs) as a decision-making component for complex autonomous driving scenarios.<n>Extensive experiments demonstrate that our proposed method not only consistently surpasses baseline approaches in single-vehicle tasks, but also helps handle complex driving behaviors even multi-vehicle coordination.
  arXiv  Detail & Related papers  (2023-10-04T17:59:49Z)
• Inverse Online Learning: Understanding Non-Stationary and Reactionary Policies [79.60322329952453]
  We show how to develop interpretable representations of how agents make decisions. By understanding the decision-making processes underlying a set of observed trajectories, we cast the policy inference problem as the inverse to this online learning problem. We introduce a practical algorithm for retrospectively estimating such perceived effects, alongside the process through which agents update them. Through application to the analysis of UNOS organ donation acceptance decisions, we demonstrate that our approach can bring valuable insights into the factors that govern decision processes and how they change over time.
  arXiv  Detail & Related papers  (2022-03-14T17:40:42Z)

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.