AI Paper Atlas

Research map / Video / Video generation

MoScale: Autoregressive Next-Scale Prediction for Human Motion Generation and Editing

Inwoo Hwang, Hojun Jang, Bing Zhou, Jian Wang, Young Min Kim, chuan guo

ECCV 2026 · Video › Video generation · approach: Text-to-motion diffusion

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 skeletal topology awareness to the tokenizer (ablation)ScaleMoGen · SnapMoGen testFID: 26.74 -> 16.35Improved
add skeletal topology awareness to the tokenizer (ablation)ScaleMoGen · SnapMoGen testCLIP Score: 0.630 -> 0.693Improved
reduce code size from 24 to 16 bits (ablation)ScaleMoGen · SnapMoGen testCLIP Score / FID: CLIP 0.693 -> 0.705; FID 16.35 -> 19.53Mixed
increase code size from 24 to 32 bits (ablation)ScaleMoGen · SnapMoGen testCLIP Score / FID: FID 16.35 -> 15.98; CLIP 0.693 -> 0.683Mixed
scale token map predictor from 125M to 2.2B (ablation)ScaleMoGen · SnapMoGen testCLIP Score: 0.693 -> 0.682Hurt
scale token map predictor from 125M to 362M (ablation)ScaleMoGen · SnapMoGen testFID: 16.35 -> 15.84No clear effect
use multi-scale skeletal-temporal bitwise VQ instead of categorical codebook (ablation)ScaleMoGen · Motion reconstructionFID: 4.83 -> 0.12Improved
remove skeletal topology from VQ (ablation)ScaleMoGen · Motion reconstructionMPJPE: 2.73 -> 6.66Improved
use ScaleMoGen next-scale prediction instead of MoMaskScaleMoGen · HumanML3D testFID: 0.045 -> 0.030Improved
use ScaleMoGen instead of MoMask++ScaleMoGen · SnapMoGen testCLIP Score: 0.685 -> 0.693Improved
use ScaleMoGen instead of MoMask++ScaleMoGen · SnapMoGen testFID: 15.06 -> 16.35Hurt
use ScaleMoGen editing instead of SALADScaleMoGen · Motion editing user studyOverall quality: 3.23 -> 4.65Improved

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