SOURCE-LINKED INTELLIGENCE

GraphAHA: Graph-Based Adaptive Search with Heterogeneous Actions for Test-Time Code Generation

arXiv · AI, language, vision and robotics · article · Sep 11, 2026 · UTC

Test-time scaling improves code generation by spending additional inference budget (e.g., calls or tokens) on direct sampling, feedback-conditioned repair, and reasoning-guided implementation. Search-based methods can allocate this budget adaptively, but two challenges remain. First, tree-structured search treats each generation history as a separate state even when trajectories converge to the same program, duplicating evaluation and preventing statistics from being shared. Second, sampling, repair, and reasoning have complementary and state-dependent payoffs, making online allocation among t

Read original source ↗ Open in workspace

recordType

paper

region

Global