{"type":"ScholarlyArticle","title":"Info-Driven Zero-Cost Proxy: Rethinking Vision Transformer Architecture Evaluation via Information Quantification","authors":[{"name":"Yue Yang","slug":"yue-yang","given":"Yue","family":"Yang","affiliation":""},{"name":"Jiacheng Wang","slug":"jiacheng-wang","given":"Jiacheng","family":"Wang","affiliation":""},{"name":"Zhenkai Yang","slug":"zhenkai-yang","given":"Zhenkai","family":"Yang","affiliation":""},{"name":"Menglan Hu","slug":"menglan-hu","given":"Menglan","family":"Hu","affiliation":""},{"name":"Gaoyang Liu","slug":"gaoyang-liu","given":"Gaoyang","family":"Liu","affiliation":""},{"name":"Bo Xu","slug":"bo-xu","given":"Bo","family":"Xu","affiliation":""},{"name":"Tianyue Zheng","slug":"tianyue-zheng","given":"Tianyue","family":"Zheng","affiliation":""},{"name":"Kai Peng","slug":"kai-peng","given":"Kai","family":"Peng","affiliation":""}],"abstract":"Neural Architecture Search (NAS) automates the design of Vision Transformer (ViT) architectures. However, the high computational cost of training-based methods has made training-free, zero-cost proxies a key research direction. While existing proxies can estimate model potential, they fail to capture critical information transmission characteristics due to the black-box nature of neural networks. Consequently, they suffer from high computational latency and weak correlation with ground-truth performance. Although information transmission determines performance, it has not been effectively quantified or utilized. This limitation remains the core bottleneck for current zero-cost proxies. In this paper, we propose Info-NAS, a zero-cost proxy based on architectural information. This method achieves the first structured quantification of information transmission in ViT. Specifically, it evaluates architectures using three core components: global information volume, local information gradient, and global consistency. Info-NAS derives proxy scores solely from architectural parameters, requiring no forward or backward propagation. Consequently, the calculation and search processes are highly efficient. Moreover, we construct the ViT-Info-Bench dataset to facilitate correlation analysis and algorithm evaluation. Experimental results demonstrate that Info-NAS significantly reduces search overhead while achieving superior ranking accuracy. With a single evaluation requiring only 1.8 ms, Info-NAS outperforms existing methods in both efficiency and performance.","doi":"10.24963/ijcai.2026/581","url":"https://ijcai.org/proceedings/2026/581","pdf":"https://ijcai.org/proceedings/2026/0581.pdf","markdown":"https://ijcai.org/proceedings/2026/581.md","bibtex":"https://ijcai.org/proceedings/2026/bibtex/581","volume":{"slug":"2026","title":"Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence","short_title":"IJCAI 2026","url":"https://ijcai.org/proceedings/2026/","year":2026,"location":"Bremen, Germany","dates":"15-21 August 2026","editors":"Diego Calvanese","isbn":"978-1-956792-09-6","publisher":"International Joint Conferences on Artificial Intelligence Organization","doi_prefix":"10.24963/ijcai.2026"},"section":"Machine Learning","track":"Main Track","number":581,"pages":{"first":5218,"last":5226},"keywords":["Machine Learning: Automated machine learning"],"videos":[],"references":[],"license_note":"Copyright International Joint Conferences on Artificial Intelligence."}