notesum.ai
Published at November 10Prompt-Efficient Fine-Tuning for GPT-like Deep Models to Reduce Hallucination and to Improve Reproducibility in Scientific Text Generation Using Stochastic Optimisation Techniques
cs.CL
cs.AI
68T50, 91B82
I.2.7; G.3
Released Date: November 10, 2024

| Model | BLEU Score | ROUGE-1 Score | ROUGE-L Score | Perplexity Score |
|---|---|---|---|---|
| GPT-2 | 0.33 | 0.42 | 0.57 | 13586.37 |
| MassSpecGPT | 0.34 | 0.44 | 0.62 | 10092.12 |