[
    {
        "id": "osp-26208",
        "type": "article-journal",
        "title": "CARE: Experience-Guided Atomic Corrective Execution for Vision-Language-Action Policies",
        "author": [
            {
                "family": "Xiao",
                "given": "Junlan"
            },
            {
                "family": "Jiang",
                "given": "Junwei"
            },
            {
                "family": "Zhang",
                "given": "Zaibin"
            },
            {
                "family": "Wang",
                "given": "Yifan"
            },
            {
                "family": "Zhang",
                "given": "Zhongbo"
            },
            {
                "family": "Lu",
                "given": "Huchuan"
            },
            {
                "family": "Wang",
                "given": "Lijun"
            }
        ],
        "URL": "https://omanscience.com/en/articles/care-experience-guided-atomic-corrective-execution-for-vision-language-action-policies",
        "language": "en",
        "issued": {
            "date-parts": [
                [
                    2026
                ]
            ]
        },
        "abstract": "Vision-Language-Action (VLA) policies achieve strong performance in robotic manipulation but remain brittle once execution deviates from nominal trajectories. We propose CARE (Corrective Atomic Robotic Execution), a framework that improves recovery by learning from failures encountered during execution. Instead of generating corrective data from manually designed or random perturbations, CARE collects failed rollouts, models stage-conditioned post-failure deviations, and uses the resulting empirical distributions to synthesize representative failure states and corrective demonstrations. At inference time, CARE combines stage-wise planning with physically grounded 3D monitoring to trigger atomic adjustments or re-operations while preserving task progress. We further introduce the Failure State Recovery Benchmark (FSR-Bench), which evaluates recovery from intermediate failure states under local deviations and structural anomalies. Experiments across multiple VLA backbones, simulation benchmarks, and real-world dual-arm tasks show consistent improvements, with average task-success gains of 14.5 points in simulation and 15.9 points in the real world. Code, models, and data are available at https://github.com/xiaojunlan/care"
    }
]