DSR
Collection
📝Official model and dataset for the paper:
Decompose, Structure, and Repair: A Neuro-Symbolic Framework for Autoformalization via Operator Trees • 2 items • Updated
YAML Metadata Warning:empty or missing yaml metadata in repo card
Check out the documentation for more information.
The DSR Formalizer was created by applying LoRA fine-tuning to Qwen2.5-7B-Instruct with the training dataset.
The instruction used for fine-tuning is as follows. To ensure model performance, please use the same instruction for inference.
Please translate the natural language component into Lean4 code, and then parse it into a structured operator tree in JSON format. Use 'formal_content' for the operator logic (with '
- Downloads last month
- 4
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support