GURU: A Hierarchical Neuro-Symbolic Multi-Agent System for Verification-Aware Automation of Mission-Critical Digital Engineering and Operations
cs.AI, cs.CE, cs.MA, cs.SY, eess.SY
Submitted: 2026-10-31
Updated: 2026-10-31
Comments: 82 pages, 50 figures, 7 tables
License: http://creativecommons.org/licenses/by/4.0/
The gist: An AI can find a safe action faster than any human, but it must never decide what counts as safe.
Terminology
Abstract
An AI can find a safe action faster than any human, but it must never decide what counts as safe. We present GURU, a hierarchical neuro-symbolic multi-agent system for mission-critical digital engineering and operations, built on one partition of authority: learned models propose, rank, and accelerate; a symbolic rules engine checks every consequential action against hard limits before it executes; and validated physics confirms and decides. GURU runs a workflow as a closed control loop, with Navigation, Action, and Planning agents sharing one Common Internal Language and one training factory. Nuclear, aviation, and national-security guidance now requires this pattern; a harnessed language model keeps the model at the center, GURU puts a validated reference there. In computational fluid dynamics, the workflow raised the converged-simulation rate on unseen aircraft geometries from under 1 percent to 74 percent over 100 single-attempt cases, ran 54 live cases to Mach 6 on three facilities, and an ablation put the symbolic layer at 13 points of boundary-layer capture. Knowledge-graph retrieval and role adapters raised ParaView control success from 55 to 75 percent. In space domain awareness, where a language model cannot be trusted to decide, a frozen self-supervised world model under validated propagation scored highest of five methods at conjunction risk, including fully fine-tuned language models, with several-times-lower seed variance: the trustworthy design gives up nothing in accuracy, and the recipe transferred to reactor anomaly detection. A radar-to-avoidance debris pipeline was demonstrated in stages on a 2026 reentry; the hierarchy coordinates simulated teams of 200 aircraft. Training scaled near-linearly past 1,000 Aurora nodes and ran at 8,000. The approach extends to a control-grade microreactor digital twin with validated physics, not a learned model, as the safety basis.
Sources
- Can A Single Human Supervise A Swarm of 100 Heterogeneous Robots?
- V-JEPA 2: Self-Supervised Video Models Enable Understanding, Prediction and Planning
- LeJEPA: Provable and Scalable Self-Supervised Learning Without the Heuristics
- Revisiting Feature Prediction for Learning Visual Representations from Video
- How to Scale Your EMA
- Decision Transformer: Reinforcement Learning via Sequence Modeling
- Run Time Assured Reinforcement Learning for Six Degree-of-Freedom Spacecraft Inspection
- Joint Embeddings Go Temporal
- AutoGluon-Tabular: Robust and Accurate AutoML for Structured Data
- The Safe Trusted Autonomy for Responsible Space Program
- LoRA: Low-Rank Adaptation of Large Language Models
- When Does LeJEPA Learn a World Model?
- Multi-Agent Actor-Critic for Mixed Cooperative-Competitive Environments
- LeWorldModel: Stable End-to-End Joint-Embedding Predictive Architecture from Pixels
- An Empirical Model of Large-Batch Training
- UMAP: Uniform Manifold Approximation and Projection for Dimension Reduction
- Evaluation of a Tree-based Pipeline Optimization Tool for Automating Data Science
- Can You Trust Your Model's Uncertainty? Evaluating Predictive Uncertainty Under Dataset Shift
- CatBoost: unbiased boosting with categorical features
- Toolformer: Language Models Can Teach Themselves to Use Tools
Related papers
- Multi-Objective Structured Pruning of LLMs for Latency and Model Size Optimization
- MAVEN-T: Reinforced Heterogeneous Distillation for Real-Time Multi-Agent Trajectory Prediction
- Model Discovery Agent: LLM-assisted Bayesian experiment design for data-efficient discovery of mechanistic world models
- Embeddings for Preferences, Not Semantics
- Can We Trust Item Response Theory for AI Evaluation?
- Diagnosing Task Insensitivity in Language Agents