Channel foundation models (CFMs) are rapidly developing, with studies indicating that pretraining offers significant benefits across downstream wireless tasks. However, CFMs are typically evaluated within model-specific pipelines that utilize varying data, radio configurations, partitions, adaptation procedures, task definitions, and metrics. Consequently, existing comparisons often show that pretraining outperforms supervised training from scratch within one pipeline but fail to rank CFMs or compare them fairly with task-specific models.
To address this gap, we introduce CFM-Bench, a unified multi-domain, multi-task benchmark. It curates six channel configurations spanning 3GPP statistical simulation, two independent ray-tracing pipelines, industrial and aerial measurements, and synchronized vehicular multimodal simulation. Official partitions isolate complete trajectories, measurement sessions, vehicle links, simulation realizations, or buffered spatial regions. CFM-Bench does not prescribe an external pretraining corpus or strategy; no benchmark split may be used for foundation-model pretraining, and the official training split is reserved exclusively for downstream fine-tuning.
The benchmark additionally requires disclosure of all data used during model development and prohibits training-stage use of official test units. Six task groups are organized along three CFM application dimensions: physical-layer (PHY) channel intelligence, radio-access-network (RAN) decision intelligence, and integrated sensing and communication (ISAC). They cover CSI feedback, frequency and temporal channel extrapolation, propagation-state classification, current- and future-beam prediction, and single-frame and temporal localization. CFM-Bench provides a common substrate for comparing the transferability of channel representations across models, domains, and tasks.
Blogger's Review: The launch of CFM-Bench marks a significant advancement in the evaluation of channel foundation models by standardizing the assessment process, thus facilitating fair comparisons among different models. This not only enhances transparency in research but also lays a solid foundation for future advancements in wireless communication technology. I look forward to seeing more researchers leverage this benchmark to drive progress in the field!