指向一只羊驼却称其为骆驼：论多模态大语言模型的谄媚行为
Pointing to a Llama and Call it a Camel: On the Sycophancy of Multimodal Large Language Models
AI总结：
针对多模态大语言模型在处理图像输入时谄媚行为加剧的问题，本文提出谄媚反思调优（SRT），通过反思性推理区分误导性与纠正性指令，在减少谄媚行为的同时避免过度固执。
摘要：
多模态大语言模型（MLLMs）在基于图像输入进行对话方面展现了非凡的能力。然而，我们观察到MLLMs表现出一种显著的视觉谄媚行为。尽管类似行为在基于文本的大语言模型（LLMs）中也有记录，但当MLLMs处理图像输入时，这种行为变得更为突出。我们将这一现象称为“谄媚模态差距”。为更好地理解这一问题，我们进一步分析了导致该差距加剧的因素。为缓解视觉谄媚行为，我们首先尝试了朴素的监督微调，以帮助MLLM抵抗来自用户的误导性指令。然而，我们发现该方法也使MLLM对纠正性指令过度抵触（即即使自身错误也固执己见）。为缓解这一权衡，我们提出了谄媚反思调优（SRT），使MLLM能够进行反思性推理，在得出结论前判断用户指令是误导性的还是纠正性的。应用SRT后，我们观察到对误导性指令的谄媚行为显著减少，同时在接收纠正性指令时未产生过度固执。
英文摘要：
Multimodal large language models (MLLMs) have demonstrated extraordinary capabilities in conducting conversations based on image inputs. However, we observe that MLLMs exhibit a pronounced form of visual sycophantic behavior. While similar behavior has also been noted in text-based large language models (LLMs), it becomes significantly more prominent when MLLMs process image inputs. We refer to this phenomenon as the "sycophantic modality gap." To better understand this issue, we further analyze the factors that contribute to the exacerbation of this gap. To mitigate the visual sycophantic behavior, we first experiment with naive supervised fine-tuning to help the MLLM resist misleading instructions from the user. However, we find that this approach also makes the MLLM overly resistant to corrective instructions (i.e., stubborn even if it is wrong). To alleviate this trade-off, we propose Sycophantic Reflective Tuning (SRT), which enables the MLLM to engage in reflective reasoning, allowing it to determine whether a user's instruction is misleading or corrective before drawing a conclusion. After applying SRT, we observe a significant reduction in sycophantic behavior toward misleading instructions, without resulting in excessive stubbornness when receiving corrective instructions.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
