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¶

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

Subclaims¶

q2 i? LongTutor Evidence averages increase from 25.67% to 65.88% for OmniEdu-4B, from 5.81% to 66.63% for OmniEdu-9B, and from 36.80% to 78.20% for OmniEdu-27B. → Hao Liang 2026 q2 i? OmniEdu-27B and OmniEdu-9B rank first and second in LongTutor knowledge-state diagnosis accuracy at 54.04% and 53.55%, but the article notes diagnosis remains comparatively challenging. → 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

Pedagogical-tutoring benchmark evaluation (Section 4.4, Table 3) on MathTutorBench, TutorBench, and LongTutor. The article reports LongTutor Evidence average gains at all three scales, indicating stronger use of student learning history.

"The Evidence average increases from 25.67% to 65.88% for OmniEdu-4B, from 5.81% to 66.63% for OmniEdu-9B, and from 36.80% to78.20%for OmniEdu-27B."

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

LongTutor knowledge-state diagnosis results in Section 4.4. The article reports OmniEdu-27B and OmniEdu-9B ranking first and second at 54.04% and 53.55%, while noting "knowledge-state diagnosis remains comparatively challenging".

"OmniEdu also performs particularly well on knowledge-state diagnosis, with OmniEdu-27B and OmniEdu-9B ranking first and second in diagnosis accuracy at54.04%and53.55%, respectively. However, the relatively low accuracies indicate that knowledge-state diagnosis remains comparatively challenging."

Discussion¶

Related Claims¶

• 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% — 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
• OmniEdu-27B achieves the best overall curriculum-grounding performance among evaluated models, raising K12-Bench EM from 52.11% to 63.12% — related
• TACTutor shows no degradation relative to its Qwen backbone on external educational benchmarks, with scores rising on MRBench, TutorBench, LongTutor, and DeepTutor — 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
• K12-Train-Full achieves the best overall performance on three multimodal educational benchmarks, and textual and visual supervision are complementary — related