[
    {
        "id": "osp-17569",
        "type": "article-journal",
        "title": "Adaptive Utilization of Low-Rank Adaptation via Conditioned Gating",
        "author": [
            {
                "family": "Yang",
                "given": "Guang"
            },
            {
                "family": "Guan",
                "given": "Changhao"
            },
            {
                "family": "Huang",
                "given": "Chao"
            },
            {
                "family": "Chen",
                "given": "Yufeng"
            },
            {
                "family": "Huang",
                "given": "Kaiyu"
            }
        ],
        "URL": "https://omanscience.com/en/articles/adaptive-utilization-of-low-rank-adaptation-via-conditioned-gating",
        "language": "en",
        "issued": {
            "date-parts": [
                [
                    2026
                ]
            ]
        },
        "abstract": "Low-Rank Adaptation (LoRA) achieves parameter-efficient fine-tuning by constraining model updates to a low-rank subspace and has been widely used in practice. However, LoRA typically employs a shared low-rank update across tokens, which limits its ability to fully exploit the adaptation subspace for tokens from different sequences. To address this issue, we propose an adaptive utilization of Low-Rank Adaptation (U-LoRA), which employs conditioned gating to explicitly learn effective token-level utilization of the limited low-rank adaptation subspace. Specifically, U-LoRA generates utilization coefficients along low-rank directions for each token and jointly coordinates and constrains them using sequence-level contextual information, thereby inducing more consistent adaptive patterns within a sentence. To further enhance training stability, we introduce a bias-corrected exponential moving average (EMA) historical prior that calibrates utilization signals across optimization steps, suppressing noise caused by batch-to-batch fluctuations. The effectiveness of our method arises from a better utilization of the existing low-rank subspace via input-conditioned strategies, rather than from expanding the subspace. Experiments on mathematical reasoning and natural language understanding benchmarks demonstrate that U-LoRA achieves competitive performance under comparable parameter budgets when with strong LoRA baselines and recent variants."
    }
]