Instructions to use ylfeng/ReF-Decompile with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use ylfeng/ReF-Decompile with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="ylfeng/ReF-Decompile")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("ylfeng/ReF-Decompile")
model = AutoModelForCausalLM.from_pretrained("ylfeng/ReF-Decompile")
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=40)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use ylfeng/ReF-Decompile with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "ylfeng/ReF-Decompile"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "ylfeng/ReF-Decompile",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/ylfeng/ReF-Decompile

SGLang

How to use ylfeng/ReF-Decompile with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "ylfeng/ReF-Decompile" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "ylfeng/ReF-Decompile",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "ylfeng/ReF-Decompile" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "ylfeng/ReF-Decompile",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use ylfeng/ReF-Decompile with Docker Model Runner:
```
docker model run hf.co/ylfeng/ReF-Decompile
```

ReF-Decompile / README.md

ylfeng

Update README.md

afbaa0c verified 7 months ago

preview code

raw

history blame contribute delete

3.53 kB

metadata

library_name: transformers
license: mit
model-index:
  - name: ref-decompile
    results: []
pipeline_tag: text-generation
datasets:
  - ylfeng/ReF-Decompile-dataset

ReF Decompile

This model is a fine-tuned version of LLM4Binary/llm4decompile-6.7b-v1.5.

Results

Model/Metrics	Re-executability Rate (%)					Readability (#)
Model/Metrics	O0	O1	O2	O3	AVG	O0	O1	O2	O3	AVG
Rule Based Decompiler
ghidra	34.76	16.46	15.24	14.02	20.12	2.98	2.41	2.52	2.38	2.57
Refine-Based Method
GPT-4o	46.95	34.15	28.66	31.10	35.22	2.82	2.35	2.29	2.31	2.44
LLM4Decompile-Ref	74.39	46.95	47.56	42.07	52.74	4.08	3.38	3.34	3.19	3.50
End-to-End Method
LLM4Decompile-End	69.51	44.51	39.63	38.41	48.02	4.07	3.46	3.40	3.23	3.54
FAE Decompile	67.68	48.78	45.73	42.07	51.07	3.94	3.46	3.40	3.25	3.51
FAE Decompile+SCC	70.24	48.54	47.56	43.29	52.41	3.97	3.48	3.41	3.23	3.52
ReF Decompile	85.37	56.10	51.83	52.43	61.43	4.13	3.60	3.54	3.49	3.69

Resources

Reference

@misc{feng2025refdecompilerelabelingfunction,
      title={ReF Decompile: Relabeling and Function Call Enhanced Decompile}, 
      author={Yunlong Feng and Bohan Li and Xiaoming Shi and Qingfu Zhu and Wanxiang Che},
      year={2025},
      eprint={2502.12221},
      archivePrefix={arXiv},
      primaryClass={cs.SE},
      url={https://arxiv.org/abs/2502.12221}, 
}