Benchmark RadarSolarChain-Eval
A physics-constrained benchmark for evaluating economic agents in decentralized energy markets, with a Gymnasium-compatible MDP and an LLM-based Planner/Auditor layer. It evaluates market utility, physical safety, slippage, action smoothness, spatial fairness, and auditability.
- Released
- 2026-07-09
- Readiness
- Paper only
- Primary field
- Cybersecurity
Why it matters
The benchmark addresses the need to evaluate agent trustworthiness in cyber-physical systems, where agents must balance utility with safety. However, without clear public access to the environment and scoring mechanisms, its standalone value is limited.
Motivation
As agentic AI systems are increasingly applied to cyber-physical environments, their evaluation requires assessment of both task performance and trustworthiness.
Primary resources
Benchmark Radar records only publicly supported details and links back to primary sources for verification.