Large language models are increasingly capable at closed-world mathematical reasoning, but research assistance also requires source-grounded use of the literature. When a proof reaches a non-trivial step, a useful assistant should determine whether the needed tool (e.g., a lemma) already exists, identify a suitable scholarly source, and verify that its assumptions align with the current proof context. To rigorously evaluate such capabilities, we introduce Re2Math, a benchmark for tool-grounded retrieval from partial mathematical proofs. Each instance is built from a candidate instrumental citation in the proof of a main theorem, with hierarchical context and an optional leakage-controlled anchor hint. We also make the task source-grounded yet citation-agnostic in that any admissible theorem sufficient for the proof transition is accepted. Evaluation uses a release-frozen retrieval artifact, ensuring reproducibility, while the benchmark itself supports automatic, continual expansion with newly constructed instances. On the current benchmark test set, the best fixed-judge ToolAcc reaches 7.0%, despite substantially higher rates of source grounding, indicating that current systems often retrieve valid statements but fail to establish their applicability to the local proof step. By decoupling citation recall, grounding, and proof-gap sufficiency, Re2Math transforms literature-grounded mathematical tool use into a controlled diagnostic task.
@article{arxiv.2605.09012,
title = {Re$^2$Math: Benchmarking Theorem Retrieval in Research-Level Mathematics},
author = {Zicheng Lyu and Wenjie Yang and Shengzhong Zhang and Zengfeng Huang},
journal= {arXiv preprint arXiv:2605.09012},
year = {2026}
}