AI Paper Atlas

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

Mantis: A Versatile Vision-Language-Action Model with Disentangled Visual Foresight

Yi Yang, Xueqi Li, Yiyang Chen, Jin Song, Yihan Wang, Zipeng Xiao, Jiadi Su, You Qiaoben, Pengfei Liu, Zhijie Deng

CVPR 2026 · Vision-language and multimodal › Robot action and navigation · approach: Vision-language-action foundation models · 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 residual connection of current frame to the DiT head (ablation)Mantis DVF · LIBERO (avg)success rate (%): 94.4 -> 95.7Improved
add the disentangled visual foresight head (ablation)Mantis · LIBERO (avg)success rate (%): 91.3 -> 95.7Improved
pretrain the DVF on human and robot videos (ablation)Mantis · LIBERO (avg)success rate (%): 95.7 -> 96.2Improved
add residual connection (Long suite) (ablation)Mantis DVF · LIBERO Longsuccess rate (%): 89.8 -> 92.2Improved
add DVF (Spatial suite) (ablation)Mantis · LIBERO Spatialsuccess rate (%): 93.4 -> 98.2Improved
replace temporal ensemble with adaptive temporal ensemble (ablation)Mantis · LIBERO (4 suites)inference count: reduces the IC by nearly 50%Improved
replace temporal ensemble with adaptive temporal ensemble (ablation)Mantis · LIBERO (4 suites)success rateNo clear effect
add language supervision (vs Mantis-LU) (ablation)Mantis · Real-world OOD instructionssuccessImproved
decouple foresight prediction from action learningMantis vs UnifiedVLA · LIBERO Spatial fine-tuningconvergence speed: success rate of zero during the first ten epochsImproved
Mantis with language supervision vs pi0.5 without, OOD scenario 1Mantis · Real-world scenario 1 OODavg successes: 3.5 -> 8.25Improved
Mantis vs pi0.5, OOD scenario 2Mantis · Real-world scenario 2 OODavg successes: 2.5 -> 7.75Improved
Mantis vs pi0.5, ID scenario 1Mantis · Real-world scenario 1 IDavg successes: 7.75 -> 8.25Mixed

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 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