Magnifier Prompt: Tackling Multimodal Hallucination via Extremely Simple Instructions
This addresses hallucinations in multimodal models for practical applications, though it is incremental as it builds on existing prompt-based approaches.
The paper tackles multimodal hallucination in large language models by proposing Magnifier Prompt (MagPrompt), a simple, training-free method using instructions to prioritize image information over internal knowledge, achieving performance comparable to or better than complex methods like VCD across multiple datasets.
Hallucinations in multimodal large language models (MLLMs) hinder their practical applications. To address this, we propose a Magnifier Prompt (MagPrompt), a simple yet effective method to tackle hallucinations in MLLMs via extremely simple instructions. MagPrompt is based on the following two key principles, which guide the design of various effective prompts, demonstrating robustness: (1) MLLMs should focus more on the image. (2) When there are conflicts between the image and the model's inner knowledge, MLLMs should prioritize the image. MagPrompt is training-free and can be applied to open-source and closed-source models, such as GPT-4o and Gemini-pro. It performs well across many datasets and its effectiveness is comparable or even better than more complex methods like VCD. Furthermore, our prompt design principles and experimental analyses provide valuable insights into multimodal hallucination.