[
    {
        "id": "osp-19879",
        "type": "article-journal",
        "title": "ReCast: Contract-Preserving Protection for Fixed-Interface Multimodal Reasoning",
        "author": [
            {
                "family": "Pei",
                "given": "Bingchen"
            },
            {
                "family": "Chen",
                "given": "Lichong"
            },
            {
                "family": "Zhao",
                "given": "Bingxi"
            },
            {
                "family": "Wu",
                "given": "Ziang"
            },
            {
                "family": "Wang",
                "given": "Sirui"
            },
            {
                "family": "Zhang",
                "given": "Min"
            },
            {
                "family": "Chen",
                "given": "Yanhao"
            },
            {
                "family": "Liu",
                "given": "Qingxu"
            },
            {
                "family": "Gao",
                "given": "Qiang"
            },
            {
                "family": "Lu",
                "given": "Chang-Tien"
            },
            {
                "family": "Gao",
                "given": "Bo"
            }
        ],
        "URL": "https://omanscience.com/en/articles/recast-contract-preserving-protection-for-fixed-interface-multimodal-reasoning",
        "language": "en",
        "issued": {
            "date-parts": [
                [
                    2026
                ]
            ]
        },
        "abstract": "Remote multimodal models offer strong numerical reasoning capabilities over charts and speech, but sending private inputs risks exposing sensitive content. Text-only sanitization cannot directly satisfy fixed media interfaces, while identity anonymization leaves the underlying task content exposed. We introduce ReCast, an agentic plug-in framework that replaces source-specific content while preserving task-relevant relations and the required input modality. ReCast locally converts inputs into a shared textual evidence-query record, jointly rewrites entities and topics with a distilled 4B model, and substitutes values through a locally invertible, role-aware numerical map. A reconstruction agent generates and validates the required media from the protected record. The remote solver returns a program whose protected operands are restored locally before execution. On 4,000 held-out ChartQA and NMSQA examples, ReCast achieves 75.10% accuracy, retaining 92.43% of unprotected remote accuracy, while a model-based audit flags source-content leakage in 7.95% of solver-bound requests. It outperforms all evaluated local baselines, preserving the benefit of remote reasoning while reducing source-content exposure under existing media interfaces."
    }
]