AI Paper Atlas

Research map / Probabilistic and scientific modeling / Differential equations and PDEs

(U)NFV: Supervised and Unsupervised Neural Finite Volume Methods for Solving Hyperbolic PDEs

Nathan Lichtlé, Alexi Canesse, Zhe Fu, HOSSEIN NICK ZINAT MATIN, Maria Laura Delle Monache, Alexandre M Bayen

ICLR 2026 · Probabilistic and scientific modeling › Differential equations and PDEs · approach: Neural PDE solvers

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
replace Godunov flux with learned NFV_3^1 fluxfinite volume solver · Burgers syntheticL2 error: 1.9e-3 -> 8.5e-4Improved
unsupervised weak-form training (UNFV) vs supervised NFV (ablation)NFV_3^1 · synthetic hyperbolic PDEsL2 error: Greenshields 1.3e-4 (NFV) vs 2.0e-4 (UNFV); both beat Godunov 4.5e-4Mixed
enlarge stencil from 3x1 to 5x5 (ablation)NFV · Burgers syntheticL2 error: 8.3e-4 -> 2.2e-4Improved
enlarge stencil from 5x5 to 11x11 (ablation)NFV · I-24 field dataL1 error: 2.31e-2 -> 2.02e-2Improved
NFV_3^1 vs ENO/WENONFV_3^1 · synthetic PDEsL2 error: Underwood 3.8e-4 vs WENO 9.8e-5 (WENO better); Greenshields 1.3e-4 vs 6.4e-4Mixed
NFV vs DGNFV_5^5 · synthetic PDEsL2 error: DG 1.0e-4 vs NFV_5^5 2.2e-4 on BurgersHurt
NFV_11^11 on unseen traffic days vs GodunovNFV_11^11 · I-24 MOTION unseen daysL2 error: 3.74e-2 -> 2.20e-2Improved
NFV_3^1 vs calibrated Godunov fits on field dataNFV_3^1 · I-24 MOTIONL1 error: 2.77e-2 -> 2.37e-2Improved

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