AI Paper Atlas

Research map / Label-efficient and robust learning / Self-supervised and contrastive

FM4NPP: A Scaling Foundation Model for Nuclear and Particle Physics

David Keetae Park, Shuhang Li, Yi Huang, Xihaier Luo, Haiwang Yu, Yeonju Go, Christopher Pinkenburg, Yuewei Lin, Shinjae Yoo, Joseph D. Osborn, Jin Huang, Yihui Ren

ICLR 2026 · Label-efficient and robust learning › Self-supervised and contrastive · approach: Self-supervised speech and brain decoding · 12 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add pretrained FM4NPP(m6) features to adapter (ablation)adapter · Track FindingARI: 0.7243 -> 0.9448Improved
add pretrained FM4NPP(m6) features to adapter (ablation)adapter · Particle Identificationaccuracy: 0.6631 -> 0.9039Improved
add pretrained FM4NPP(m6) features to adapter (ablation)adapter · Noise Taggingaccuracy: 0.9111 -> 0.9713Improved
FM4NPP with 2.39M-parameter adapter versus Exa.TrkX GNNExa.TrkX · Track Findingpurity (%): 66.42 -> 93.08Improved
FM4NPP versus OneFormer3D on noise tagging recallOneFormer3D · Noise Taggingmacro recall: 0.9404 -> 0.9367No clear effect
increase pretrained FM size (ablation)FM4NPP · downstream tasksperformanceImproved
increase FM size from m5 to m6 (ablation)FM4NPP · pretraining validation MSEvalidation MSENo clear effect
increase pretraining dataset size (ablation)FM4NPP m3 · pretraining validation MSEvalidation MSEImproved
pretrained FM with few labeled examples versus many (ablation)FM4NPP · Track Findingrelative ARI gain: 2.9x vs 1.3xImproved
FM4NPP versus sPHENIX Cellular Automaton + Kalman pipelinesPHENIX pipeline · Track Finding (high-pT tracks)tracking efficiency (%): 94.6 -> 99.6Improved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy