Researchers at USTC have developed a new AI system called ‘Woodpecker’ that can detect and correct hallucinated text generated by multimodal language models (MLLM) when describing images.
View Article on VentureBeat
AI,Business,Data Infrastructure,Programming & Development,AI, ML and Deep Learning,Conversational AI,MLLM,MLLMs,multimodal large language model,multimodal models,NLP,Tencent,USTC,Woodpecker
Programming & Development