On the RiceChem ALAG grading benchmark GLM-4-9B-chat leads six LLMs (Accuracy 0.6492, F1 0.6054)¶

Claim · All claims
Evidence · 1 study · 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? Evaluating six LLMs on the RiceChem long-answer grading task, GLM-4-9B-chat leads across Accuracy, Precision, Recall, and F1 (Accuracy = 0.6492, F1 = 0.6054), with Mistral-7B-Instruct-v0.2 ranking second (Accuracy = 0.6167, F1 = 0.5681). → Fuan Xiao 2026

Evidence¶

Fuan Xiao 2026¶

Fuan Xiao, Jiahui Huang, Jia-Xin Huang, Huali Ren and Lang Li. (2026). Integrating LLM with consortium blockchain for personalized and verifiable online education in higher education. International Journal of Educational Technology in Higher Education. https://doi.org/10.1186/s41239-026-00618-5

q2 · i? · design · r2

Benchmark evaluation (Table 2) following Sonkar et al.'s RiceChem ALAG long-answer setting, measuring Accuracy, Precision, Recall, and F1 for six LLMs on the same dataset. GLM-4-9B-chat achieved the leading "Accuracy = 0.6492, F1 = 0.6054".

"Overall, GLM-4-9B-chat leads across metrics (Accuracy = 0.6492, F1 = 0.6054), indicating a strong balance between precision and recall; Mistral-7B-Instruct-v0.2 ranks second (Accuracy = 0.6167, F1 = 0.5681) with similarly robust aggregate performance."

Discussion¶

Related Claims¶

• LLM graders show model-specific precision/recall trade-offs, so a single metric is insufficient to judge grading quality — related