Can a Lightweight Automated AI Pipeline Solve Research-Level Mathematical Problems?

ArXi:2602.13695v2 Announce Type: replace Large language models (LLMs) have recently achieved remarkable success in generating rigorous mathematical proofs, with "AI for Math" emerging as a vibrant field of research (Ju, 2026). While these models have mastered competition-level benchmarks like the International Mathematical Olympiad (Huang, 2025; Duan, 2025) and show promise in research applications through auto-formalization (Wang, 2025), their deployment via lightweight, natural-language pipelines for research problems remains underexplored.