# SEval-NAS: A Search-Agnostic Evaluation for Neural Architecture Search

Source: [SEval-NAS: A Search-Agnostic Evaluation for Neural Architecture Search](<https://www.sota2.com/research/paper/69a66bdc2d603e6de3e3cc9a>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-03-04T05:00:05.078Z

Authors: Atah Nuh Mih, Jianzhou Wang, Truong Thanh Hung Nguyen, Hung Cao

Published: 2026-02-17T15:02:02.000Z

## Abstract

Neural architecture search (NAS) automates the discovery of neural networks that meet specified criteria, yet its evaluation procedures are often hardcoded, limiting the ability to introduce new metrics. This issue is especially pronounced in hardware-aware NAS, where objectives depend on target devices such as edge hardware. To address this limitation, we propose SEval-NAS, a metric-evaluation mechanism that converts architectures to strings, embeds them as vectors, and predicts performance metrics. Using NATS-Bench and HW-NAS-Bench, we evaluated accuracy, latency, and memory. Kendall's $\\tau$ correlations showed stronger latency and memory predictions than accuracy, indicating the suitability of SEval-NAS as a hardware cost predictor. We further integrated SEval-NAS into FreeREA to evaluate metrics not originally included. The method successfully ranked FreeREA-generated architectures, maintained search time, and required minimal algorithmic changes. Our implementation is available at: https://github.com/Analytics-Everywhere-Lab/neural-architecture-search

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [Image Classification on NATS-Bench CIFAR-10 TSS (test)](<https://www.sota2.com/research/sota/image-classification-on-nats-bench-cifar-10-tss-test.md>): Image Classification; NATS-Bench CIFAR-10 TSS (test); Accuracy: 94.36.

- [Image Classification on NATS-Bench CIFAR-100 TSS (test)](<https://www.sota2.com/research/sota/image-classification-on-nats-bench-cifar-100-tss-test.md>): Image Classification; NATS-Bench CIFAR-100 TSS (test); Accuracy: 73.51.

- [Image Classification on NATS-Bench ImageNet16-120 TSS (test)](<https://www.sota2.com/research/sota/image-classification-on-nats-bench-imagenet16-120-tss-test.md>): Image Classification; NATS-Bench ImageNet16-120 TSS (test); Accuracy: 46.34.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2603.00099>)

[PDF](<https://arxiv.org/pdf/2603.00099.pdf>)