AI Paper Atlas

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

NeurVLA: Unleashing Failure-Handling Capability of Vision-Language-Action Models via Neural-Symbolic Reasoning

Xuqi Liu, Minghe Gao, Juncheng Li, Siliang Tang

ICML 2026 · Vision-language and multimodal › Robot action and navigation · approach: Vision-language-action foundation models

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 observation-grounded failure correction (COG) (ablation)InstructVLA · SimplerEnv-WidowXsuccess rate avg (%): 44.9 -> 55.4Improved
add trajectory-modeling failure prevention (PTM) (ablation)InstructVLA · SimplerEnv-WidowXsuccess rate avg (%): 44.9 -> 52.8Improved
combine COG and PTM (ablation)InstructVLA · SimplerEnv-WidowXsuccess rate avg (%): 44.9 -> 64.7Improved
replace SFT with reasoning-guided action learning (LRG) on top of COG (ablation)InstructVLA · SimplerEnv-WidowXsuccess rate avg (%): 55.4 -> 62.5Improved
replace SFT with LRG on top of PTM (ablation)InstructVLA · SimplerEnv-WidowXsuccess rate avg (%): 52.8 -> 59.1Improved
add LRG to COG + PTM (full NeurVLA) (ablation)InstructVLA · SimplerEnv-WidowXsuccess rate avg (%): 64.7 -> 72.8Improved
use program-derived instead of LLM-direct correction ECoTs (ablation)InstructVLA with NeurVLA · SimplerEnv-WidowXsuccess rate avg (%): 57.2 -> 72.8Improved
use program-derived instead of LLM-direct prevention ECoTs (ablation)InstructVLA with NeurVLA · SimplerEnv-WidowXsuccess rate avg (%): 62.5 -> 72.8Improved
use program-derived instead of LLM-direct ECoTs for both (ablation)InstructVLA with NeurVLA · SimplerEnv-WidowXsuccess rate avg (%): 50.4 -> 72.8Improved
apply NeurVLA to NORANORA · LIBEROavg success rate (%): 87.9 -> 96.1Improved
apply NeurVLA to InstructVLAInstructVLA · SimplerEnvavg success rate (%): 62.3 -> 83.8Improved
apply NeurVLA to NORANORA · CALVIN ABC->Davg completed length: 3.60 -> 4.09Improved

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