Re:Form -- Reducing Human Priors in Scalable Formal Software Verification with RL in LLMs: A Preliminary Study on Dafny
Paper
•
2507.16331
•
Published
•
16
Re:Form
The resources of Re:Form - Reducing Human Priors in Scalable Formal Software Verification with RL in LLMs.
-
-
Veri-Code/ReForm-Python2Dafny-Dataset
Viewer • Updated • 17.5k • 36 -
Veri-Code/ReForm-DafnyComp-Benchmark
Viewer • Updated • 300 • 35 -
Veri-Code/ReForm-14B-RL-entropy
Text Generation • 15B • Updated • 9 • 2 -
Veri-Code/ReForm-SFT-14B
Text Generation • 15B • Updated • 11 • 1 -
Veri-Code/ReForm-SFT-7B
Text Generation • 8B • Updated • 10 • 1 -
Veri-Code/ReForm-SFT-3B
Text Generation • 3B • Updated • 8 • 1 -
Veri-Code/ReForm-SFT-1.5B
Text Generation • 2B • Updated • 9 -
Veri-Code/ReForm-SFT-0.5B
Text Generation • 0.5B • Updated • 9