AI Paper Atlas

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

ForceVLA2: Unleashing Hybrid Force-Position Control with Force Awareness for Contact-Rich Manipulation

Yang Li, Zhaxizhuoma Zhaxizhuoma, Hongru Jiang, Junjie Xia, Hongquan Zhang, Jinda Du, Yunsong Zhou, Jia Zeng, Ce Hao, Jieji Ren, Qiaojun Yu, Cewu Lu and 2 more

CVPR 2026 · Vision-language and multimodal › Robot action and navigation · approach: Memory and subgoal-based manipulation · 26 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 force prompt to π0 (ablation)π0 · ForceVLA2-Dataset avgsuccess rate (%): 18.0 -> 27.0Improved
add multimodal encoder on top of force prompt (ablation)π0 + force prompt · ForceVLA2-Dataset avgsuccess rate (%): 27.0 -> 40.0Improved
add Cross-Scale MoE (ablation)π0 + force prompt + multimodal encoder · ForceVLA2-Dataset avgsuccess rate (%): 40.0 -> 66.0Improved
add force modality input to Cross-Scale MoE (ablation)ForceVLA2 · ForceVLA2-Dataset avgsuccess rate (%): 50.0 -> 66.0Improved
add visual modality input to Cross-Scale MoE (ablation)ForceVLA2 · ForceVLA2-Dataset avgsuccess rate (%): 36.0 -> 50.0Improved
add extra force token at MoE input (ablation)ForceVLA2 · Press bottle / Clean boardsuccess rate (%): 85.0 -> 80.0 (Press bottle), 85.0 -> 70.0 (Clean board)Hurt
naively concatenate force input to π0π0 · ForceVLA2-Dataset avgsuccess rate (%): 18.0 -> 17.0No clear effect
use ForceVLA2 instead of ForceVLAcontact-rich VLA · ForceVLA2-Dataset avgsuccess rate (%): 35.0 -> 66.0Improved
use ForceVLA2 instead of π0.5contact-rich VLA · ForceVLA2-Dataset avgsuccess rate (%): 31.0 -> 66.0Improved

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