AI Paper Atlas

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

Transolver-3: Scaling Up Transformer Solvers to Industrial-Scale Geometries

Hang Zhou, Haixu Wu, Haonan Shangguan, Yuezhou Ma, Huikun Weng, Jianmin Wang, Mingsheng Long

ICML 2026 · Probabilistic and scientific modeling › Differential equations and PDEs · approach: Neural PDE solvers · 0 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
use Transolver-3 instead of AB-UPTNeural PDE solver · NASA-CRM / AhmedML / DrivAerMLrelative L2 error: DrivAerML u 5.93 -> 4.14; NASA-CRM ps 9.77 -> 8.71Improved
use Transolver-3 instead of AB-UPT on AhmedML volume pressureNeural PDE solver · AhmedMLvolume pressure pv error: 2.07 -> 2.16Hurt
use Transolver-3 on Aircraft dataset with shocksTransolver++ / AB-UPT · Aircraftsurface pressure error: 6.40 -> 5.80 (Transolver++); 6.93 -> 5.80 (AB-UPT)Improved
reorder Physics-Attention with faster slice and deslice (move projections into slice domain) (ablation)Physics-Attention · Single-GPU capacitycells per GPU / latency: about 10% more cells than Transolver++; latency about 60% lower; GFLOPs 20% lowerImproved
add geometry slice tiling (ablation)Transolver-3 · Single-GPU capacitymax cells: boosts capacity by roughly 90% (1.9x Transolver++, 2.9M cells)Improved
reduce tile size from 800k to 100k to 5k (ablation)Geometry slice tiling · 800k input trainingmemory / latency: memory 22.16 -> 12.28 -> 11.04 GB; latency 404 -> 429 -> 1307 msMixed
increase training subset size n from 1k to 100k (ablation)Geometry amortized training · DrivAerMLrelative L2 error: ps 4.84 -> 3.71; pv 7.94 -> 5.72Improved
train on small random subset (n=1000) (ablation)Geometry amortized training · DrivAerMLrelative L2 errorNo clear effect
use more cells for physical state cache up to full mesh (ablation)Physical state caching · DrivAerMLrelative L2 errorImproved
increase evaluation surface cell density (ablation)Full mesh decoding · DrivAerMLdrag and lift coefficient errorImproved
use decoupled inference compared with AB-UPTTransolver-3 · DrivAerMLinference memory / latency: memory 11.2 -> 6.0 GB; latency 220 -> 260 sMixed

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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