Less is More: Mitigating Multimodal Hallucination from an EOS Decision Perspective
Paper • 2402.14545 • Published
How to use yuezih/llava-v1.5-7b-selective-23k-lora with PEFT:
from peft import PeftModel
from transformers import AutoModelForCausalLM
base_model = AutoModelForCausalLM.from_pretrained("llava-v1.5-7b")
model = PeftModel.from_pretrained(base_model, "yuezih/llava-v1.5-7b-selective-23k-lora")Official LoRA weights of paper Less is More: Mitigating Multimodal Hallucination from an EOS Decision Perspective.