AI Paper Atlas

Research map / Vision-language and multimodal / CLIP and zero-shot

Interleave-VLA: Enhancing Robot Manipulation with Image-Text Interleaved Instructions

Cunxin Fan, Xiaosong Jia, Yihang Sun, Yixiao Wang, Jianglan Wei, Ziyang Gong, Xiangyu Zhao, Masayoshi Tomizuka, Xue Yang, Junchi Yan, Mingyu Ding

ICLR 2026 · Vision-language and multimodal › CLIP and zero-shot · approach: Open-vocabulary alignment from VLMs

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
train and test with interleaved image-text instructions instead of textπ0 · SimplerEnv Novel Objectsuccess rate: 26.0 ± 3.6 -> 53.8 ± 1.5Improved
train and test with interleaved image-text instructions instead of textπ0 · SimplerEnv Novel Categorysuccess rate: 19.3 ± 1.5 -> 57.3 ± 2.8Improved
train with interleaved data but test with text only (ablation)π0 · SimplerEnv Novel Objectsuccess rate: 26.0 ± 3.6 -> 35.8 ± 0.2Improved
interleaved instructions on familiar tasksπ0 · SimplerEnv In-Domainsuccess rate: 68.1 ± 1.3 -> 70.5 ± 1.3No clear effect
pretrain on Interleaved X-Embodiment dataset (ablation)Interleave-VLA · Real robot FANUC Out-of-Domainsuccess rate: 0 -> 71Improved
interleaved instructions instead of text, both pretrainedπ0 · Real robot FANUC Out-of-Domainsuccess rate: 13 -> 71Improved
mix Internet and task-specific prompt images (ablation)Interleave-VLA · SimplerEnv In-Domainsuccess rate: 59.2 -> 71.0 (vs Internet only)Improved
mix Internet and task-specific prompt images (ablation)Interleave-VLA · SimplerEnv Out-of-Domainsuccess rate: 67.1 -> 71.7 (vs task-specific only)Improved
add interleaved format text to visual goal images (ablation)Interleave-VLA · SimplerEnv Move Nearsuccess rate: 0.0 -> 68.8Improved
add visual goal image content to text instruction (ablation)π0 · SimplerEnv Unseen Cat.success rate: 21.0 -> 54.2Improved
apply Interleave-VLA to OpenVLA with InternVL2.5OpenVLA · VIMA-Bench L1success rate: 53.71 -> 83.14Improved
swap OpenVLA VLM backbone Prismatic to InternVL2.5 (text input) (ablation)OpenVLA · VIMA-Bench L1-L3success rate: 53.71 -> 45.29 (L1), 23.00 -> 24.71 (L2), 23.86 -> 29.71 (L3)No clear effect

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