Papers

Trending research and the full catalog - each paper linked to the benchmarks, methods, and models it introduces.

Filtered by domain: mathClear

FlowBalance: Verifier-Grounded Self-Improvement from On-Policy Reasoning Experience

3 Sep 2026

A reasoning model can improve from its own on-policy experience, but this inner loop is fragile: terminal verifiers provide reliable yet sparse supervision, while dense same-model guidance can reinforce false confidence or overconcentrate learning on a narrow solution mode.

Mathematics Reasoning

Filtering here covers the 2,000 most recent papers, as much as one page can hold in memory. See the full index of 22,029 papers.