ZODIAC: Zero-shot Offline Diffusion for Inferring Multi-xApps Conflicts in Open Radio Access Networks

2026-04-21

AI Summary

This work addresses the challenge of detecting conflicts among concurrently executing multi-vendor xApps in Open Radio Access Networks (O-RAN), which are difficult to identify during development and for which existing methods rely on scarce co-execution data. To overcome this limitation, the paper introduces ZODIAC, a novel framework that enables zero-shot conflict inference using only offline marginal data from individual xApps. ZODIAC employs a three-stage pipeline—uncertainty-aware proxy modeling, trajectory-level diffusion training, and composition-guided denoising—augmented with physics-informed constraints and offline reinforcement learning. The approach establishes a theoretical lower bound for compositional conflict inference that effectively proxies real-world conflict severity. Experiments on Mobile-Env and NS-O-RAN-Flexric demonstrate that ZODIAC improves Top-20 true positive rate by over 20% and significantly enhances Spearman rank correlation, while maintaining high scenario diversity and computational efficiency.

Technology Category

• Constraint Satisfaction and Optimization: Distributed CSP/Optimization
• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems
• Search and Optimization: Applications

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Security and Privacy: Data transparency and provenance

Abstract

Open Radio Access Network (O-RAN) enables network control through multi-vendor xApps operating both within and across layers, subnets, and domains, whose concurrent execution can trigger conflicts that are latent during the development phase. Existing conflict management approaches rely heavily on joint-execution data, which is often unavailable in practice. To address this limitation, we formalize a novel problem termed conflict reasoning, which involves identifying conflict-inducing conditions given only marginal datasets from each individual xApp. We propose ZODIAC, a three-stage framework for zero-shot conflict condition inference that comprises uncertainty-aware surrogate model training, trajectory-level diffusion training, and compositional guided denoising for efficient, physics-constrained, and reliable condition search. We derive a theoretical lower confidence bound showing that the compositional reasoning in ZODIAC serves as a principled surrogate for true conflict severity, with the epistemic penalty directly controlling the approximation gap. We evaluate ZODIAC on both the lightweight Mobile-Env platform across all three O-RAN Alliance conflict types (direct, indirect, and implicit) and a realistic NS-O-RAN-Flexric simulator. ZODIAC consistently outperforms baseline condition search methods, achieving over 20% higher True Positive Rate at Top-20, substantially stronger Spearman rank correlation, greater scenario diversity, and competitive computational efficiency. Ablation studies confirm the necessity of each guidance component, with epistemic uncertainty penalties proving essential for filtering spurious conflicts. To the best of our knowledge, ZODIAC is the first framework in O-RAN that enables conflict reasoning from marginal offline data without requiring any joint-execution traces.

Problem

• O-RAN
• xApps conflicts
• zero-shot inference
• offline data
• conflict reasoning

Innovation

• zero-shot conflict reasoning
• offline diffusion
• O-RAN xApps
• compositional guided denoising
• epistemic uncertainty