通过鲁棒指令微调缓解大型多模态模型中的幻觉
Mitigating Hallucination in Large Multi-Modal Models via Robust Instruction Tuning
AI总结：
本文提出首个包含正负指令的视觉指令微调数据集LRV-Instruction，并设计GAVIE评估方法，通过微调MiniGPT4和mPLUG-Owl有效缓解了大型多模态模型的幻觉问题，同时提升了多数据集性能。
摘要：
尽管在多模态任务上取得了可喜的进展，当前的大型多模态模型（LMMs）仍容易产生与相关图像和人类指令不一致的幻觉描述。本文通过引入首个大型且多样化的视觉指令微调数据集，名为Large-scale Robust Visual (LRV)-Instruction，来解决这一问题。我们的数据集包含由GPT4生成的40万条视觉指令，涵盖16个视觉-语言任务，具有开放式的指令和答案。与现有主要关注正指令样本的研究不同，我们设计LRV-Instruction以包含正负指令，以实现更鲁棒的视觉指令微调。我们的负指令设计在三个语义层级上：(i) 不存在对象操纵，(ii) 存在对象操纵，(iii) 知识操纵。为高效衡量LMMs产生的幻觉，我们提出GPT4-Assisted Visual Instruction Evaluation (GAVIE)，这是一种稳定的评估视觉指令微调的方法，如同人类专家一样。GAVIE不需要人工标注的真实答案，并能适应多种指令格式。我们进行了全面实验来研究LMMs的幻觉。结果表明，现有LMMs在面对我们的负指令时表现出显著的幻觉，尤其是存在对象和知识操纵指令。此外，我们通过在LRV-Instruction上微调MiniGPT4和mPLUG-Owl成功缓解了幻觉，同时与最先进方法相比在多个公共数据集上提升了性能。此外，我们观察到训练数据中正负实例的平衡比例能产生更鲁棒的模型。代码和数据可在https://github.com/FuxiaoLiu/LRV-Instruction获取。
英文摘要：
Despite the promising progress in multi-modal tasks, current large multi-modal models (LMMs) are prone to hallucinating inconsistent descriptions with respect to the associated image and human instructions. This paper addresses this issue by introducing the first large and diverse visual instruction tuning dataset, named Large-scale Robust Visual (LRV)-Instruction. Our dataset comprises 400k visual instructions generated by GPT4, covering 16 vision-and-language tasks with open-ended instructions and answers. Unlike existing studies that primarily focus on positive instruction samples, we design LRV-Instruction to include both positive and negative instructions for more robust visual instruction tuning. Our negative instructions are designed at three semantic levels: (i) Nonexistent Object Manipulation, (ii) Existent Object Manipulation and (iii) Knowledge Manipulation. To efficiently measure the hallucination generated by LMMs, we propose GPT4-Assisted Visual Instruction Evaluation (GAVIE), a stable approach to evaluate visual instruction tuning like human experts. GAVIE does not require human-annotated groundtruth answers and can adapt to diverse instruction formats. We conduct comprehensive experiments to investigate the hallucination of LMMs. Our results demonstrate existing LMMs exhibit significant hallucinations when presented with our negative instructions, particularly Existent Object and Knowledge Manipulation instructions. Moreover, we successfully mitigate hallucination by finetuning MiniGPT4 and mPLUG-Owl on LRV-Instruction while improving performance on several public datasets compared to state-of-the-art methods. Additionally, we observed that a balanced ratio of positive and negative instances in the training data leads to a more robust model. Code and data are available at https://github.com/FuxiaoLiu/LRV-Instruction.
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。
