LLM-Based Agentic Negotiation for 6G: Addressing Uncertainty Neglect and Tail-Event Risk

2025-11-24

AI Summary

6G autonomous networks suffer from an “uncertainty neglect bias”: LLM-driven agents tend to optimize for mean performance while disregarding tail risks—particularly extreme latency events. To address this, we propose the first agent negotiation framework integrating digital twin technology with Conditional Value-at-Risk (CVaR), pioneering the application of Extreme Value Theory (EVT) to 6G network negotiation. Our approach enables precise modeling and control of tail latency risk by leveraging digital twins for real-time end-to-end latency distribution forecasting, quantifying tail risk via CVaR, and incorporating a cognitive uncertainty propagation mechanism to realize genuinely risk-aware negotiation. Evaluated in an eMBB/URLLC network slicing coordination scenario, the framework achieves zero SLA violations and reduces p99.999 latency by 11%, markedly enhancing system reliability. This work establishes a novel paradigm for trustworthy, risk-conscious 6G autonomy.

Technology Category

• Multiagent Systems: Agreement, Argumentation & Negotiation
• Planning, Routing, and Scheduling: Scheduling under Uncertainty
• Machine Learning: Calibration & Uncertainty Quantification

Application Category

• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Search and Retrieval-Augmented AI: Agentic search
• Social Networks and Social Media: Generative AI / large language models and their impact on social systems

Abstract

A critical barrier to the trustworthiness of sixth-generation (6G) agentic autonomous networks is the uncertainty neglect bias; a cognitive tendency for large language model (LLM)-powered agents to make high-stakes decisions based on simple averages while ignoring the tail risk of extreme events. This paper proposes an unbiased, risk-aware framework for agentic negotiation, designed to ensure robust resource allocation in 6G network slicing. Specifically, agents leverage Digital Twins (DTs) to predict full latency distributions, which are then evaluated using a formal framework from extreme value theory, namely, Conditional Value-at-Risk (CVaR). This approach fundamentally shifts the agent's objective from reasoning over the mean to reasoning over the tail, thereby building a statistically-grounded buffer against worst-case outcomes. Furthermore, our framework ensures full uncertainty awareness by requiring agents to quantify epistemic uncertainty -- confidence in their own DTs predictions -- and propagate this meta-verification to make robust decisions, preventing them from acting on unreliable data. We validate this framework in a 6G inter-slice negotiation use-case between an eMBB and a URLLC agent. The results demonstrate the profound failure of the biased, mean-based baseline, which consistently fails its SLAs with a 25% rate. Our unbiased, CVaR-aware agent successfully mitigates this bias, eliminating SLA violations and reducing the URLLC and eMBB p99.999 latencies by around 11%. We show this reliability comes at the rational and quantifiable cost of slightly reduced energy savings to 17%, exposing the false economy of the biased approach. This work provides a concrete methodology for building the trustworthy autonomous systems required for 6G.

Problem

• Addressing uncertainty neglect bias in LLM-powered 6G autonomous networks
• Mitigating tail-event risks in high-stakes network resource allocation decisions
• Ensuring robust negotiation between network slices under extreme latency conditions

Innovation

• Using Digital Twins to predict latency distributions
• Applying Conditional Value-at-Risk for tail risk evaluation
• Quantifying epistemic uncertainty for robust decision-making