# Kintsugi-VLA: Turning Failed Robot Rollouts into Recovery Data through Interventional Recoverability

arXiv: https://arxiv.org/abs/2609.31048 · HTML: https://mlsanity.com/paper/2609.31048/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Failure recovery and robustness, one of 13 directions in Vision-language-action models.
- Builds on 13 of its 15 references that are in the map.
- First posted on arXiv 25 Sep 2026.

## Placed in

- Vision-language-action models / [Failure recovery and robustness](https://mlsanity.com/vla/failure-recovery-and-robustness/)

## Builds on (most cited first, 13 of 13)

- [OpenVLA: An Open-Source Vision-Language-Action Model](https://mlsanity.com/paper/2406.09246/) (2024)
- [SmolVLA: A Vision-Language-Action Model for Affordable and Efficient Robotics](https://mlsanity.com/paper/2506.01844/) (2025)
- [MimicGen: A Data Generation System for Scalable Robot Learning using Human Demonstrations](https://mlsanity.com/paper/2310.17596/) (2023)
- [Time-uniform, nonparametric, nonasymptotic confidence sequences](https://mlsanity.com/paper/1810.08240/) (2018)
- [SAFE: Multitask Failure Detection for Vision-Language-Action Models](https://mlsanity.com/paper/2506.09937/) (2025)
- [RACER: Rich Language-Guided Failure Recovery Policies for Imitation Learning](https://mlsanity.com/paper/2409.14674/) (2024)
- [DextrAH-G: Pixels-to-Action Dexterous Arm-Hand Grasping with Geometric Fabrics](https://mlsanity.com/paper/2407.02274/) (2024)
- [RaC: Robot Learning for Long-Horizon Tasks by Scaling Recovery and Correction](https://mlsanity.com/paper/2509.07953/) (2025)
- [IntervenGen: Interventional Data Generation for Robust and Data-Efficient Robot Imitation Learning](https://mlsanity.com/paper/2405.01472/) (2024)
- [Back to the Familiar Future: Failure Recovery for VLA Policies via Pre-Imagined Milestone Selection](https://mlsanity.com/paper/2606.09258/) (2026)
- [FLARE: A Failure-Aware Framework for Autonomous Correction and Recovery in Visual-Language Robotic Manipulation](https://mlsanity.com/paper/2608.26645/) (2026)
- [Imagining Recovery: Inference-Time Counterfactual Realignment for Vision-Language-Action Models](https://mlsanity.com/paper/2608.14822/) (2026)
- [LIBERO-RECOVER: Beyond Task Success Towards Failure Recovery in Robotic Manipulation Models](https://mlsanity.com/paper/2609.05178/) (2026)

## Abstract (arXiv)

Simulation enables scalable training of Vision-Language-Action policies by using privileged experts to generate visual demonstrations without requiring every trajectory to be collected through manual teleoperation. However, such pipelines typically retain successful demonstrations while failed rollouts are discarded, even though they expose precisely the off-nominal states from which recovery must be learned. We introduce Kintsugi-VLA, a framework for converting failed rollouts into targeted synthetic recovery data by exploiting exact state restoration and branching in simulation. For a fixed privileged expert, we define interventional recoverability as the probability of completing the original task after the simulator is restored to a given state, estimate it using adaptive Monte Carlo continuations with pointwise Wilson intervals, and characterize its non-monotonic evolution along failed trajectories. These estimates identify an observed terminal low-recoverability frontier-the point after which measured recoverability remains below a threshold-which is then used to select informative recovery starting states. In a simulated Franka manipulation task, targeted recovery data yield aggregate SmolVLA recovery success of 34.6% and 38.4% under difficulty- and frame-budget matching, respectively, 5.8 and 6.7 percentage points above uniform sampling within the same recovery window. The same ordering is observed under disturbed end-to-end execution and shifted clutter and physics conditions, while clean-task success decreases from 76.8% to 74.7%. Kintsugi-VLA demonstrates how failed simulator rollouts can be transformed from discarded experience into structured recovery-training data through direct interventional measurement.