{"slug": "landslidefm-a-foundation-model-towards-unified-landslide-perception-and", "title": "LandslideFM: A Foundation Model Towards Unified Landslide Perception and Understanding", "summary": "Researchers posted arXiv:2611.12936v1, introducing LandslideFM, which they describe as the first unified foundation model connecting visual landslide perception tasks with contextual interpretation. LandslideFM uses a Source-Agreement Graduated Trust semi-supervised learning framework to combine limited labeled data with abundant unlabeled data, and feeds predicted landslide regions as explicit spatial guidance to a Vision-Language Model adapted through parameter-efficient fine-tuning. The team also released LandslideVL, described as the first benchmark for intelligent landslide analysis combining visual imagery with new textual prompts, and reported that LandslideFM surpasses general-purpose VLMs in specialized landslide reasoning and scene analysis, with code at github.com/wangshanwen001/LandslideFM.", "body_md": "arXiv:2611.12936v1 Announce Type: new \nAbstract: Intelligent landslide detection is essential for disaster prevention. However, existing models primarily focus on isolated visual tasks, require extensive labeled data, and lack deep analytical and scene-understanding capabilities. To address these limitations, we propose LandslideFM, the first unified foundation model that connects visual perception tasks with contextual interpretation of landslides. Specifically, LandslideFM adapts to multiple visual tasks via a Source-Agreement Graduated Trust semi-supervised learning framework that leverages limited labeled data alongside abundant unlabeled data. Furthermore, predicted landslide regions provide explicit spatial guidance to a Vision-Language Model adapted through parameter-efficient fine-tuning, enabling comprehensive landslide reasoning. To support training and evaluation in this domain, we also present LandslideVL, the first benchmark for intelligent landslide analysis, combining visual imagery and newly introduced textual prompts. Comprehensive experiments across public datasets and LandslideVL demonstrate that LandslideFM achieves superior quantitative and qualitative performance. Most importantly, LandslideFM surpasses general-purpose VLMs in specialized landslide reasoning and scene analysis. Our code is released at https://github.com/wangshanwen001/LandslideFM.", "url": "https://wpnews.pro/news/landslidefm-a-foundation-model-towards-unified-landslide-perception-and", "canonical_source": "https://arxiv.org/abs/2611.12936", "published_at": "2026-11-11 05:00:00+00:00", "updated_at": "2026-11-11 05:22:41.229187+00:00", "lang": "en", "topics": ["artificial-intelligence", "machine-learning", "computer-vision", "ai-research", "natural-language-processing"], "entities": ["LandslideFM", "LandslideVL", "arXiv:2611.12936v1", "Vision-Language Model", "GitHub"], "also_reported_by": [], "alternates": {"html": "https://wpnews.pro/news/landslidefm-a-foundation-model-towards-unified-landslide-perception-and", "markdown": "https://wpnews.pro/news/landslidefm-a-foundation-model-towards-unified-landslide-perception-and.md", "text": "https://wpnews.pro/news/landslidefm-a-foundation-model-towards-unified-landslide-perception-and.txt", "jsonld": "https://wpnews.pro/news/landslidefm-a-foundation-model-towards-unified-landslide-perception-and.jsonld"}}