AI Paper Atlas

Research map / Vision-language and multimodal / Robot action and navigation

ContextFlow: In-Context Flow Matching for Robot Manipulation

Jian Ding, Xianjie DAI, Roei Herzig, Nussair Hroub, Jinjie Mai, Dengxin Dai, Bernard Ghanem, Mohamed Elhoseiny

ECCV 2026 · Vision-language and multimodal › Robot action and navigation · approach: Multimodal prompt-driven manipulation · 2 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 autoregressive action decoding with flow matching (both pi0-initialized)ContextAR vs ContextFlow-Plain · LIBERO unseen configssuccess rate (%): 53.5 -> 64.0Improved
add multimodal context compressors (ablation)ContextFlow-Plain · LIBERO unseen configssuccess rate (%): 64.0 -> 73.5Improved
add only image compressor (ablation)ContextFlow-Plain · LIBERO unseen configssuccess rate (%): 64.0 -> 68.0 (Spatial 55.0 -> 51.0, Object 73.0 -> 85.0)Mixed
include proprio states and actions in context (ablation)ContextFlow · LIBERO unseen configssuccess rate (%): 29.0 -> 73.5Improved
include images in context (ablation)ContextFlow · LIBERO unseen configssuccess rate (%): 45.0 -> 73.5Improved
include text instruction in context (ablation)ContextFlow · LIBERO unseen configssuccess rate (%): 64.5 -> 73.5Improved
increase demo frames from 4 to 8 (ablation)ContextFlow · LIBERO unseen configssuccess rate (%): 70.5 -> 73.5Improved
increase states/actions from 128 to 256 (ablation)ContextFlow · LIBERO unseen configssuccess rate (%): 73.5 -> 66.5Hurt
use flow matching head instead of MLP L1 regression (ablation)ContextFlow · LIBERO unseen configssuccess rate (%): 58.0 -> 73.5Improved
use flow matching head instead of diffusion head (ablation)ContextFlow · LIBERO unseen configssuccess rate (%): 33.5 -> 73.5Improved
MLP L1 continuous parallel head vs autoregressive (ablation)ContextAR · LIBERO unseen configssuccess rate (%): 53.5 -> 58.0Improved
flow matching reduces cumulative error vs autoregressionContextAR · Cumulative MSE on S2 taskMSE: 4.7 -> 3.0Improved

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