Education-oriented tuning improves authentic K-12 problem solving, with OmniEdu-27B's MDK12-Bench open-question correctness rising from 62.08% to 78.77%¶

Claim · All claims
Evidence · 1 study (2 entries) · design r2 · q2
Currency · newest study 2026 · not yet checked against its sources · not reviewed by a person · recheck by 2029: tool and model accuracy changes quickly

Subclaims¶

q2 i? On MDK12-Bench, OmniEdu-27B improves the overall score from 46.04% to 57.76%, with open-question correctness rising from 62.08% to 78.77%. → Hao Liang 2026 q2 i? On EXAMS-V, the improvement is particularly pronounced for OmniEdu-4B, whose overall accuracy increases from 44.69% to 57.62%. → Hao Liang 2026 (2)

Evidence¶

Hao Liang 2026¶

Hao Liang, Qihan Lin, Meiyi Qiang, Linzhuang Sun, Hengyi Feng, Mingrui Chen, Sizhe Qiu, Wentao Zhang. (2026). OmniEdu: Open Foundation Models for Learning and Teaching. https://arxiv.org/abs/2609.23088 doi:10.48550/arxiv.2609.23088

q2 · i? · design · r2

K-12 problem-solving benchmark evaluation (Section 4.3, Table 2) on GAOKAO-Bench, EXAMS-V, and MDK12-Bench. The article reports the MDK12-Bench overall score rising from 46.04% to 57.76% with open-question correctness "rising substantially from 62.08% to78.77%".

"On MDK12-Bench, OmniEdu-27B improves the overall score from 46.04% to57.76%, with open-question correctness rising substantially from 62.08% to78.77%."

Hao Liang 2026 (2)¶

Hao Liang, Qihan Lin, Meiyi Qiang, Linzhuang Sun, Hengyi Feng, Mingrui Chen, Sizhe Qiu, Wentao Zhang. (2026). OmniEdu: Open Foundation Models for Learning and Teaching. https://arxiv.org/abs/2609.23088

q2 · i? · design · r2

EXAMS-V evaluation within the Section 4.3 problem-solving results. The article states the improvement is "particularly pronounced for OmniEdu-4B", with overall accuracy increasing from 44.69% to 57.62%.

"On EXAMS-V, the improvement is particularly pronounced for OmniEdu-4B, whose overall accuracy increases from 44.69% to 57.62%."

Discussion¶

Related Claims¶

• OmniEdu-27B achieves the best overall curriculum-grounding performance among evaluated models, raising K12-Bench EM from 52.11% to 63.12% — related
• Specializing models for K-12 education does not require a broad loss of general capability, with MMMU-Pro accuracy increasing at all three scales — related
• Education-oriented instruction tuning consistently improves curriculum grounding, K-12 problem solving, and pedagogical tutoring at every evaluated model scale — a broader claim this one bears on
• K12-Train-Full achieves the best overall performance on three multimodal educational benchmarks, and textual and visual supervision are complementary — related
• Education-oriented tuning produces the largest gains in scaffolded instruction and use of longitudinal student evidence, with LongTutor Evidence averages rising sharply at all scales — related