ResearchBenchmarksPuzzle Solving on Sudoku (final-step)Follow79.5Success RateActFocus13.25230.45147.6564.849May 14, 2026Evaluation ResultsMethodMethodLinksSuccess RateActFocusAlgorithm=GRPO, Model...Algorithm=GRPO, Model scale=1.5B, Evaluation step=2002026.0579.5UniformAlgorithm=GRPO, Model...Algorithm=GRPO, Model scale=1.5B, Evaluation step=2002026.0515.8