[
    {
        "id": "osp-25303",
        "type": "article-journal",
        "title": "Visual Jev: Accurate and Efficient Decisions from Shared Visual Context",
        "author": [
            {
                "family": "Yu",
                "given": "Guanxu"
            },
            {
                "family": "Yao",
                "given": "Yuhang"
            }
        ],
        "URL": "https://omanscience.com/ar/articles/visual-jev-accurate-and-efficient-decisions-from-shared-visual-context",
        "language": "en",
        "issued": {
            "date-parts": [
                [
                    2026
                ]
            ]
        },
        "abstract": "Many vision applications ask several independent, forced-choice questions about the same image. Visual Jev encodes the image and public context once, executes isolated question suffixes as a batch, and reads candidate probabilities from the backbone's language-model head. Across four benchmarks, answer-supervised post-training raises equal-weight macro accuracy from 70.6% to 76.1%, with the gain concentrated on the two task families represented in training. At N=32 questions per image, shared batched execution is 8.9x faster in warm amortized time than independent serial execution and remains 3.4x faster than an already-batched baseline that recomputes the prefix, at the cost of higher peak memory. A matched typed-head control offers no consistent accuracy advantage over the language-model-head readout. The supported design is therefore simple: adapt the backbone for quality, retain the existing readout, and share execution for efficiency."
    }
]