The ICLR 2026 best paper reveals that Transformers achieve extreme succinctness—encoding complex concepts with exponentially fewer symbols than RNNs—while proving that analyzing or verifying such models incurs EXPSPACE‑complete computational costs.
A new shortest‑path algorithm developed by researchers at Tsinghua University claims to overcome the long‑standing sorting bottleneck of Dijkstra’s classic method, extending to both undirected and directed graphs and sparking fresh debate on algorithmic optimality and future research directions.
This article presents an overview of graph neural networks, explains their computational framework, analyzes space and time complexities, and proposes ten practical strategies—including edge avoidance, dimensionality reduction, selective iteration, memory baking, distillation, partitioning, sparse computation, routing, and cross-sample feature sharing—to significantly lower the cost of large‑scale GNN processing.
