HalDet-LLaVA
HalDet-LLaVA is designed for multimodal hallucination detection, trained on the MHaluBench training dataset, achieving detection performance close to that of using GPT4-Vision.
HalDet-LLaVA is trained on the MHaluBench training set using LLaVA-v1.5, specific parameters can be found in the file finetune_task_lora.sh.
We trained HalDet-LLaVA on 1-A800 in 1 hour. If you don"t have enough GPU resources, we will soon provide model distributed training scripts.
You can inference our HalDet-LLaVA by using inference.py
To view more detailed information about HalDet-LLaVA and the train dataset, please refer to the EasyDetect and readme
- Downloads last month
- 4
Install from pip and serve model
# Install vLLM from pip: pip install vllm# Start the vLLM server: vllm serve "zjunlp/HalDet-llava-13b"# Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "zjunlp/HalDet-llava-13b", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'