无需训练即可清晰感知：减轻多模态大语言模型在遥感中的幻觉
Seeing Clearly without Training: Mitigating Hallucinations in Multimodal LLMs for Remote Sensing
AI总结：
RADAR通过利用多模态大语言模型的内在注意力，无需训练即可减少遥感视觉问答中的事实性和逻辑性幻觉。
摘要：
多模态大语言模型（MLLMs）在遥感视觉问答（RS-VQA）中表现出显著的幻觉问题，主要由大规模场景中的视觉定位失败或对细粒度小目标的误判引起。为了系统分析这些问题，我们引入RSHBench，一个基于协议的基准测试，用于对事实性和逻辑性幻觉进行细粒度诊断。为了缓解由定位引起的事实性幻觉，我们进一步提出相对注意力驱动的主动推理（RADAR），一种无需训练的推理方法，利用MLLMs内在的注意力来指导测试时的逐步定位和细粒度局部推理。在多样化的MLLMs上进行的广泛实验表明，RADAR一致提高了RS-VQA性能，并减少了事实性和逻辑性幻觉。代码和数据将在：https://github.com/MiliLab/RADAR上公开。
英文摘要：
Multimodal large language models (MLLMs) suffer from pronounced hallucinations in remote sensing visual question-answering (RS-VQA), primarily caused by visual grounding failures in large-scale scenes or misinterpretation of fine-grained small targets. To systematically analyze these issues, we introduce RSHBench, a protocol-based benchmark for fine-grained diagnosis of factual and logical hallucinations. To mitigate grounding-induced factual hallucinations, we further propose Relative Attention-Driven Actively Reasoning (RADAR), a training-free inference method that leverages intrinsic attention in MLLMs to guide progressive localization and fine-grained local reasoning at test time. Extensive experiments across diverse MLLMs demonstrate that RADAR consistently improves RS-VQA performance and reduces both factual and logical hallucinations. Code and data will be publicly available at: https://github.com/MiliLab/RADAR
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。
