HallE-Control：控制大型多模态模型中的物体幻觉
HallE-Control: Controlling Object Hallucination in Large Multimodal Models
AI总结：
本文提出CCEval评估方法和HallE-Control可控模型，通过分离上下文知识与参数化知识，在保持物体覆盖率的同时将大型多模态模型的物体存在性幻觉降低44%。
摘要：
当前的大型多模态模型（LMMs）取得了显著进展，但它们在准确理解视觉细节（即执行详细图像描述）方面的能力仍存在很大的不确定性。为解决这一问题，我们提出了CCEval，一种由GPT-4辅助的详细图像描述评估方法。有趣的是，尽管LMMs在现有的视觉问答基准中表现出极少的物体存在性幻觉，但我们提出的评估方法揭示出它们仍然容易产生此类幻觉。在本文中，我们首次尝试从不同方面研究这种幻觉，包括图像分辨率、语言解码器规模以及指令数据的数量、质量和粒度。我们的研究结果强调，当语言描述包含比视觉模块能够定位或验证的更细物体粒度的细节时，会产生无根据的推断，从而诱发幻觉。为了控制此类幻觉，我们进一步将图像描述的可靠性归因于上下文知识（仅涉及上下文中有依据的物体）和参数化知识（包含模型推断出的物体）。因此，我们提出了HallE-Control，一个在物体存在性幻觉方面可控的LMM。HallE-Control可以将图像描述条件化，使其在（i）仅描述有依据物体的上下文知识，与（ii）将其与参数化知识融合以想象推断出的物体之间切换。与LLaVA$_{7B}$相比，我们的方法将幻觉减少了44%，同时保持了物体覆盖率。
英文摘要：
Current Large Multimodal Models (LMMs) achieve remarkable progress, yet there remains significant uncertainty regarding their ability to accurately apprehend visual details, that is, in performing detailed captioning. To address this, we introduce $\textit{CCEval}$, a GPT-4 assisted evaluation method for detailed captioning. Interestingly, while LMMs demonstrate minimal object existence hallucination in existing VQA benchmarks, our proposed evaluation reveals continued susceptibility to such hallucinations. In this paper, we make the first attempt to investigate such hallucination from different aspects, including image resolution, the language decoder size, and instruction data amount, quality, granularity. Our findings underscore the unwarranted inference when the language description includes details at a finer object granularity than what the vision module can ground or verify, thus inducing hallucination. To control such hallucinations, we further attribute the reliability of captioning to contextual knowledge (involving only contextually grounded objects) and parametric knowledge (containing inferred objects by the model). Thus, we introduce $\textit{HallE-Control}$, a controllable LMM in terms of $\textbf{Hall}$ucination in object $\textbf{E}$xistence. HallE-Control can condition the captioning to shift between (i) exclusively depicting contextual knowledge for grounded objects and (ii) blending it with parametric knowledge to imagine inferred objects. Our method reduces hallucination by 44% compared to LLaVA$_{7B}$ and maintains the object coverage.
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。
