{"slug": "a-map-of-all-9110-neurips-2026-papers-qwen3-embedding-umap-datamapplot", "title": "A map of all 9,110 NeurIPS 2026 papers (Qwen3-Embedding + UMAP + DataMapPlot)", "summary": "Hadi Mohammadi embedded all 9,110 NeurIPS 2026 paper titles and abstracts with Qwen3-Embedding-0.6B, clustered them into 150 topics via k-means, and laid them out with UMAP and DataMapPlot on one interactive map, with 2025 papers assigned to the same topics to show growth and decline. LLM agent papers rose from 2.9% to 6.8% of the conference, and 6 of the 10 fastest-growing topics concern agents, including testing, reliability, memory and safety. The map, search interface and code are published on Mohammadi's site and GitHub.", "body_md": "NeurIPS 2026 has 9,110 papers. I embedded all of them (title + abstract) with Qwen3-Embedding-0.6B, grouped them into 150 topics with k-means, laid them out with UMAP and put everything on one interactive map with DataMapPlot. The 2025 papers go into the same topics, so the map can be coloured by what grew and what shrank.\n\nThe biggest change: LLM agent papers went from 2.9% to 6.8% of the conference, and 6 of the 10 fastest-growing topics are about agents (testing them, reliability, memory, safety).\n\nMap and search: [What 9,110 NeurIPS 2026 papers are about — Hadi Mohammadi](https://mohammadi.cv/neurips/2026/)\n\nCode: [GitHub - mohammadi-hadi/neurips: NeurIPS Atlas: what NeurIPS papers are about and how the conference changes. 2026 edition: 9,110 papers on one map, growth since 2025, three cities, and every paper with its links. · GitHub](https://github.com/mohammadi-hadi/neurips)\n\nFeedback on the topic names is very welcome.", "url": "https://wpnews.pro/news/a-map-of-all-9110-neurips-2026-papers-qwen3-embedding-umap-datamapplot", "canonical_source": "https://discuss.huggingface.co/t/a-map-of-all-9-110-neurips-2026-papers-qwen3-embedding-umap-datamapplot/190859#post_1", "published_at": "2026-10-21 08:35:33+00:00", "updated_at": "2026-10-21 08:43:22.403367+00:00", "lang": "en", "topics": ["ai-research", "large-language-models", "ai-agents", "machine-learning"], "entities": ["Hadi Mohammadi", "NeurIPS 2026", "Qwen3-Embedding-0.6B", "UMAP", "DataMapPlot", "GitHub"], "also_reported_by": [], "alternates": {"html": "https://wpnews.pro/news/a-map-of-all-9110-neurips-2026-papers-qwen3-embedding-umap-datamapplot", "markdown": "https://wpnews.pro/news/a-map-of-all-9110-neurips-2026-papers-qwen3-embedding-umap-datamapplot.md", "text": "https://wpnews.pro/news/a-map-of-all-9110-neurips-2026-papers-qwen3-embedding-umap-datamapplot.txt", "jsonld": "https://wpnews.pro/news/a-map-of-all-9110-neurips-2026-papers-qwen3-embedding-umap-datamapplot.jsonld"}}