{"slug": "closed-loop-control-with-rule-aligned-small-language-models-and-multi-agent-self", "title": "Closed-Loop Control with Rule-Aligned Small Language Models and Multi-Agent Self-Correction", "summary": "Researchers developed a closed-loop control framework using a small language model (Qwen2.5-1.5B) aligned via GRPO and a multi-agent self-correction loop with a digital-twin validator. In thermal-control simulations, the system achieved 91.5% action-alignment accuracy at 3.84s mean inference latency, demonstrating a practical path for reconfigurable autonomous control at the edge.", "body_md": "arXiv:2607.09713v1 Announce Type: new\nAbstract: A key step toward autonomous industrial operation is the ability to create and reconfigure control policies from natural-language requirement specifications, with minimal or no manual redesign. In this setting, policy generation by AI agents can be a credible path when paired with a plant-aware validator (e.g., a digital twin) that can check generated candidate actions before execution. However, practical deployment is constrained by inference latency and compute footprint: large cloud-based models are often too slow, opaque, or data-sensitive for edge closed-loop use. This work investigates whether a compact Small Language Model (SLM) can be retrained for control reasoning and embedded in a validator-guided correction loop. We use a Qwen2.5-1.5B model aligned via Group Relative Policy Optimization (GRPO), combined with (i) an action agent, (ii) a symbolic/digital-twin-style validation layer, and (iii) a reprompting agent that iteratively steers outputs toward valid actions. In randomized thermal-control simulations (30 experiments with 500 steps each), the framework achieves 91.5% average action-alignment accuracy (86.3%--100% across cases) at 3.84\\,s mean inference latency. Under symbolic re-mapping, it maintains a 95% in-range rate, indicating robust physical regulation despite reduced token-level agreement. These results support SLM+validator architectures as a practical path toward reconfigurable autonomous control at the edge.", "url": "https://wpnews.pro/news/closed-loop-control-with-rule-aligned-small-language-models-and-multi-agent-self", "canonical_source": "https://arxiv.org/abs/2607.09713", "published_at": "2026-07-14 04:00:00+00:00", "updated_at": "2026-07-14 04:23:54.087608+00:00", "lang": "en", "topics": ["artificial-intelligence", "large-language-models", "ai-agents", "ai-infrastructure", "ai-research"], "entities": ["Qwen2.5-1.5B", "Group Relative Policy Optimization"], "alternates": {"html": "https://wpnews.pro/news/closed-loop-control-with-rule-aligned-small-language-models-and-multi-agent-self", "markdown": "https://wpnews.pro/news/closed-loop-control-with-rule-aligned-small-language-models-and-multi-agent-self.md", "text": "https://wpnews.pro/news/closed-loop-control-with-rule-aligned-small-language-models-and-multi-agent-self.txt", "jsonld": "https://wpnews.pro/news/closed-loop-control-with-rule-aligned-small-language-models-and-multi-agent-self.jsonld"}}