AI Paper Atlas

Research map / Video / Video generation

DisCoRD: Discrete Tokens to Continuous Motion via Rectified Flow Decoding

Jungbin Cho, Junwan Kim, Jisoo Kim, Minseo Kim, Mingu Kang, Sungeun Hong, Tae-Hyun Oh, Youngjae Yu

ICCV 2025 · Video › Video generation · approach: Text-to-motion diffusion · 16 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
replace feed-forward decoder with DisCoRD rectified flow decoderMoMask · HumanML3D reconstructionsJPE: 0.512 -> 0.385Improved
replace decoder with DisCoRDMoMask · HumanML3D text-to-motionFID: 0.045 -> 0.032Improved
replace decoder with DisCoRDT2M-GPT · HumanML3D text-to-motionR-Precision Top 3: 0.775 -> 0.760Hurt
replace decoder with DisCoRDMoMask · HumanML3D reconstructionMPJPE: 29.5 -> 33.3Hurt
replace decoder with DisCoRD on TM2DTM2D · AIST++ music-to-danceFID_g: 20.09 -> 88.74Hurt
use DisCoRD conditioned decoding instead of rectified flow post refinement (ablation)MoMask · HumanML3D text-to-motionFID: 0.035 -> 0.032Improved
use repeat-stack-project condition projection instead of up-convolution (ablation)DisCoRD · HumanML3D text-to-motionFID: 0.039 -> 0.032Improved
add attention to U-Net backbone (ablation)DisCoRD · HumanML3D text-to-motionFID: 0.032 -> 0.043Hurt
train on sliding windows instead of full motion sequences (ablation)DisCoRD · HumanML3D text-to-motionFID: 0.038 -> 0.032Improved
increase residual quantization level of the tokenizer (ablation)MoMask + DisCoRD · HumanML3D text-to-motionR-Precision gain: RP@1 -2.8% at R0 -> +0.6% at R5Improved
use rectified flow with ODE sampler instead of diffusion DDIM at 2 steps (ablation)DisCoRD decoder · HumanML3D decodingFID: 0.055 -> 0.034 (2 steps)Improved
use sJPE instead of MPJPE for human-perceived naturalnessevaluation metric · user study correlationPearson correlation with human score: 0.181 -> 0.483Improved

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