Abstract
Large Multimodal Models (LMMs) have achieved impressive progress in visual perception and reasoning. However, when confronted with visually ambiguous or non-semantic scene text, they often struggle to accurately spot and understand the content, frequently generating semantically plausible yet visually incorrect answers, which we refer to as semantic hallucination. In this work, we investigate the underlying causes of semantic hallucination and identify a key finding: Transformer layers in LLM with stronger attention focus on scene text regions are less prone to producing semantic hallucinations. Thus, we propose a training-free semantic hallucination mitigation framework comprising two key components: (1) ZoomText, a coarse-to-fine strategy that identifies potential text regions without external detectors; and (2) Grounded Layer Correction, which adaptively leverages the internal representations from layers less prone to hallucination to guide decoding, correcting hallucinated outputs for non-semantic samples while preserving the semantics of meaningful ones. To enable rigorous evaluation, we introduce TextHalu-Bench, a benchmark of over 1,730 samples spanning both semantic and non-semantic cases, with manually curated question-answer pairs designed to probe model hallucinations. Extensive experiments demonstrate that our method not only effectively mitigates semantic hallucination but also achieves strong performance on public benchmarks for scene text spotting and understanding.
Abstract (translated)
大型多模态模型（LMMs）在视觉感知和推理方面取得了显著进展。然而，面对视觉模糊或非语义场景文本时，它们往往难以准确识别和理解内容，并常常生成语义上看似合理但实际上与图像不符的答案，我们称之为语义幻觉。在这项工作中，我们探讨了导致语义幻觉的潜在原因并发现了一个关键点：大型语言模型（LLM）中注意力更强的Transformer层在处理场景文本区域时较少出现语义幻觉。基于这一发现，我们提出了一种无训练方法来减轻语义幻觉框架，该框架包含两个核心组件： 1. **ZoomText**：一种从粗到细策略，无需外部检测器即可识别潜在的文字区域。 2. **Grounded Layer Correction**（地面层校正）：此模块会根据模型中较不易产生幻觉的层数内部表示来调整解码过程，在不损害有意义样本语义的情况下纠正非语义样本的输出。 为了严格评估该方法的效果，我们引入了TextHalu-Bench，这是一个包含超过1,730个样本的基准测试集，这些样本涵盖了从有语义到无语义的各种场景，并配有人工策划的问题-答案对，旨在探测模型可能出现的幻觉。广泛的实验表明，我们的方法不仅有效减轻了语义幻觉问题，在公共场景文本检测和理解基准上也取得了很强的表现。
URL
https://arxiv.org/abs/2506.05551
