[
    {
        "id": "osp-16015",
        "type": "article-journal",
        "title": "G-CARB: Graph-Localized Conformal Agent Risk Budget for Compositional Harm",
        "author": [
            {
                "family": "Yu",
                "given": "Zijun"
            },
            {
                "family": "Gu",
                "given": "Yu"
            },
            {
                "family": "Nia",
                "given": "Vahid Partovi"
            },
            {
                "family": "Asgharian",
                "given": "Masoud"
            }
        ],
        "URL": "https://omanscience.com/en/articles/g-carb-graph-localized-conformal-agent-risk-budget-for-compositional-harm",
        "language": "en",
        "issued": {
            "date-parts": [
                [
                    2026
                ]
            ]
        },
        "abstract": "Small language model (SLM) agents need safety controls that track consequences across tool calls with little monitoring overhead. A private read, for example, becomes a leak when a later action sends that data outside the system. We introduce CARB (Conformal Agent Risk Budget), which calibrates when to stop an agent using a ledger of harm incurred before stopping. Under exchangeable episodes, standard conformal risk control bounds this declared loss in expectation over calibration and a future episode. G-CARB selects scorer evidence along observable dependencies from private sources to outgoing actions. The ledger still covers the entire executed history, and computing the gate score requires no additional language-model inference. On AgentDojo replay with two 14B backbones, G-CARB roughly halves scorer-input records at intermediate risk budgets while improving autonomous task completion relative to full-prefix scoring; random context of the same size achieves similar gains. Controlled examples show how retaining the relevant dependency can further avoid stopping benign work."
    }
]