Cross-prompt pre-finetuning shows no significant benefit when sufficient (200-answer) in-prompt training data is available¶

Claim · All claims
Evidence · 1 study · causal r2 · q2
Currency · newest study 2025 · not yet checked against its sources · not reviewed by a person

Subclaims¶

q2 i? There was no significant improvement in performance when 200 answers were used for training. → Funayama 2025

Evidence¶

Funayama 2025¶

Funayama, H., Matsubayashi, Y., Asazuma, Y., Mizumoto, T., & Inui, K. (2025). Cross-prompt Pre-finetuning of Language Models for Short Answer Scoring. Int J Artif Intell Educ. https://doi.org/10.1007/s40593-025-00474-w

q2 · i? · causal · r2

In the BERT experiment varying finetuning data size (10, 25, 50, 100, 200 instances), the article reports no statistically significant difference at 200 answers; equivalence was not tested. The article states pre-finetuning "is not beneficial when sufficient training data are available."

"However, there was no significant improvement in performance when 200 answers were used for training, indicating that pre-finetuning is not beneficial when sufficient training data are available."

Discussion¶

Related Claims¶

• Cross-prompt finetuning of a pre-finetuned 8B LLM on 50 instances per prompt achieves average QWK of 0.85, comparable to professional human graders (0.87) — related
• Two-phase cross-prompt pre-finetuning with key phrases significantly improves short answer scoring accuracy when in-prompt training data is limited — related
• Pre-finetuning performance increases with the number of distinct prompts used, holding total pre-finetuning answers fixed — related
• Pre-finetuning without key phrases does not improve scoring performance and slightly lowers it relative to the BERT baseline — related