AI Paper Atlas

Research map / Probabilistic and scientific modeling / Variational and generative sampling

Terminal Velocity Matching

Linqi Zhou, Mathias Parger, Ayaan Haque, Jiaming Song

ICLR 2026 · Probabilistic and scientific modeling › Variational and generative sampling · approach: Flow matching and flow sampling · 14 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
Match terminal velocity (TVM) instead of MeanFlow initial-velocity matchingDiT-XL/2 one-step generator · ImageNet-256 1-NFEFID: 3.43 -> 3.29Improved
Use TVM with 4 NFEs instead of 500-NFE DiTImageNet-256 · ImageNet-256FID: 2.27 -> 1.99Improved
Use TVM with 4 NFEs vs DiT on ImageNet-512ImageNet-512 · ImageNet-512FID: 3.04 -> 2.94Improved
Sample constant CFG instead of random CFG during training (ablation)TVM · ImageNet-256 1-NFE, w=2FID: 5.14 -> 4.81Improved
Use EMA target (gamma=0.99) instead of no EMA (ablation)TVM pseudo-target · ImageNet-256 1-NFEFID: 10.24 -> 4.90Improved
Add scaled parameterization with CFG weight (w=1.5) (ablation)TVM · ImageNet-256 1-NFEFID: 9.32 -> 6.04Improved
Add scaled parameterization (w=2) (ablation)TVM · ImageNet-256FID 1-NFE / 2-NFE: 1-NFE 3.82 -> 3.72; 2-NFE 3.27 -> 3.35Mixed
Train with t=s (pure FM) samples 10-30% of the time (ablation)TVM · ImageNet-256FID 1-NFE / 2-NFE: 1-NFE 3.72 -> 3.91; 2-NFE 3.35 -> 2.97 (20%)Mixed
Sample t biased toward 1 and s toward 0 (truncated sampling) (ablation)TVM time sampling · ImageNet-256 1-NFEFID: 4.59 -> 3.70Improved
Over-bias time sampling (mu_t=2.0, sigma_t=2.0) (ablation)TVM time sampling · ImageNet-256 1-NFEFID: 3.70 -> 7.88 (varied configs)Hurt
Use gap sampling with lower mean gap (ablation)TVM time sampling · ImageNet-256 1-NFEFID: 5.12 -> 3.72Improved
Use higher CFG weight in training (ablation)TVM · ImageNet-256FID 1-NFE vs 2-NFEMixed

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