Research

Paper

AI LLM March 03, 2026

Seeing Clearly without Training: Mitigating Hallucinations in Multimodal LLMs for Remote Sensing

Authors

Yi Liu, Jing Zhang, Di Wang, Xiaoyu Tian, Haonan Guo, Bo Du

Abstract

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

Metadata

arXiv ID: 2603.02754
Provider: ARXIV
Primary Category: cs.CV
Published: 2026-03-03
Fetched: 2026-03-04 03:41

Related papers

Raw Data (Debug)
{
  "raw_xml": "<entry>\n    <id>http://arxiv.org/abs/2603.02754v1</id>\n    <title>Seeing Clearly without Training: Mitigating Hallucinations in Multimodal LLMs for Remote Sensing</title>\n    <updated>2026-03-03T08:53:20Z</updated>\n    <link href='https://arxiv.org/abs/2603.02754v1' rel='alternate' type='text/html'/>\n    <link href='https://arxiv.org/pdf/2603.02754v1' rel='related' title='pdf' type='application/pdf'/>\n    <summary>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</summary>\n    <category scheme='http://arxiv.org/schemas/atom' term='cs.CV'/>\n    <published>2026-03-03T08:53:20Z</published>\n    <arxiv:primary_category term='cs.CV'/>\n    <author>\n      <name>Yi Liu</name>\n    </author>\n    <author>\n      <name>Jing Zhang</name>\n    </author>\n    <author>\n      <name>Di Wang</name>\n    </author>\n    <author>\n      <name>Xiaoyu Tian</name>\n    </author>\n    <author>\n      <name>Haonan Guo</name>\n    </author>\n    <author>\n      <name>Bo Du</name>\n    </author>\n  </entry>"
}