AI Paper Atlas

Research map / Reinforcement learning / Visual and representation RL

ModSkill: Physical Character Skill Modularization

Yiming Huang, Zhiyang Dou, Lingjie Liu

ICCV 2025 · Reinforcement learning › Visual and representation RL · approach: Dexterous manipulation transfer · 7 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
add per-body-part modular low-level controllers (ablation)ModSkill · AMASS-Test full-body trackingE_g-mpjpe (mm): 41.1 -> 35.9Improved
add skill modularization attention layer (ablation)ModSkill · AMASS-Test full-body trackingE_g-mpjpe (mm): 35.9 -> 32.4Improved
add generative adaptive sampling (ablation)ModSkill · AMASS-Test full-body trackingE_mpjpe (mm): 23.2 -> 22.7Improved
add generative adaptive sampling (ablation)ModSkill · AMASS-Test full-body trackingsuccess rate: 99.3% -> 99.3%No clear effect
use ModSkill instead of PHC+ for trackingAMASS-Test · AMASS-Test full-body trackingE_g-mpjpe (mm): 36.1 -> 32.2Improved
use five body parts instead of two (ablation)ModSkill · AMASS-TestE_g-mpjpe (mm): 36.4 -> 32.2Improved
use modular skills instead of distilled PULSE embeddings for downstream reachModSkill · Downstream reachnormalized return: 0.77 -> 0.79No clear effect
apply part-wise adversarial priors to ASE (ASE-PMP)ASE · Downstream steeringnormalized return: 0.60 -> 0.31Hurt
use ModSkill instead of PULSE in VR trackingAMASS-Test VR tracking · AMASS-Test VR trackingsuccess rate: 93.4% -> 93.4%No clear effect
use ModSkill instead of PHC+ on AMASS-Train successAMASS-Train · AMASS-Train full-body trackingsuccess rate: 100% -> 99.6%Hurt

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