AI Paper Atlas

Research map / Reinforcement learning / Visual and representation RL

Sample from What You See: Visuomotor Policy Learning via Diffusion Bridge with Observation-Embedded Stochastic Differential Equation

Zhaoyang Liu, Mokai Pan, Zhongyi Wang, Kaizhen Zhu, Haotao Lu, Haipeng Zhang, Jingya Wang, Ye Shi

ICML 2026 · Reinforcement learning › Visual and representation RL · approach: Skill learning and diffusion policies · 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
replace diffusion bridge head with standard diffusion (ablation)BridgePolicy · Adroit Doorsuccess rate: 0.68 -> 0.84Improved
replace regression head with diffusion bridge (ablation)BridgePolicy · Adroit Doorsuccess rate: 0.00 -> 0.84Improved
use diffusion bridge instead of diffusion (ablation)BridgePolicy · Adroit Pensuccess rate: 0.43 -> 0.58Improved
use cross-attention instead of concatenation for fusion (ablation)BridgePolicy · Adroit Doorsuccess rate: 0.59 -> 0.665Improved
use cross-attention instead of concatenation for fusion (ablation)BridgePolicy · MetaWorld Handle-Pullsuccess rate: 0.55 -> 0.63Improved
add contrastive alignment loss (alpha 0.0 to 1.0) (ablation)BridgePolicy · DexArt Laptopsuccess rate: 0.85 -> 0.91Improved
add alignment loss alpha 0.5 (ablation)BridgePolicy · MetaWorld Coffee-Pushsuccess rate: 0.91 -> 0.99Improved
raise alignment weight alpha from 1.0 to 5.0 (ablation)BridgePolicy · MetaWorld Pick-Place-Wallsuccess rate: 0.96 -> 0.83Hurt
sample FlowPolicy with 10 steps instead of 1 (ablation)FlowPolicy · 7 simulation tasks (avg)success rate: 0.64 -> 0.59Hurt
L1 objective instead of L2 for diffusion bridge training (ablation)BridgePolicy · Adroit (3 tasks avg)success rate: 0.80 -> 0.81No clear effect
increase steady variance lambda from 30 to 50 (gamma 10^7) (ablation)BridgePolicy · Adroit (3 tasks avg)success rate: 0.77 -> 0.87Improved
increase penalty gamma from 10^5 to 10^7 (lambda 50) (ablation)BridgePolicy · Adroit (3 tasks avg)success rate: 0.83 -> 0.87Improved

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