{"ID":5937605,"CreatedAt":"2026-07-07T03:14:33.014478982Z","UpdatedAt":"2026-07-08T05:08:25.597675483Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2607.04118","arxiv_id":"2607.04118","title":"Parametric Memory Decoding for Zero-Shot Routing in LoRA-Based External Parametric Memory","abstract":"With the rise of parametric memory, LoRA-based External Parametric Memory (EPM) has emerged as a modular solution, but existing routing methods often introduce additional training, deployment, and maintenance overhead. This raises a natural question: can a LoRA-based EPM bank be routed without maintaining an additional routing component? However, existing zero-shot LoRA routing methods still face two problems under the EPM setting: (1) their evaluations are scattered across different task settings rather than organized around EPM access, and (2) their routing signals lack a unified perspective to guide systematic improvement. To address these problems, we organize PMD-Bench, covering document-level, domain-level knowledge, and task-skill, and propose Parametric Memory Decoding (PMD), the first framework designed to systematically improve zero-shot LoRA routing by reframing it as decoding activations over external parametric memory. Based on PMD, we further instantiate PMDRouter, which scores each LoRA by its response magnitude from a single base-model prefill. Experiments on PMD-Bench show that PMDRouter achieves the strongest internal-signal performance across multiple zero-shot routing settings. These results demonstrate the feasibility of zero-shot LoRA routing and suggest that PMD can serve as a general framework for improving zero-shot routing methods. Sources: Github (https://anonymous.4open.science/r/Parametric-Memory-Decoding-872A/)","short_abstract":"With the rise of parametric memory, LoRA-based External Parametric Memory (EPM) has emerged as a modular solution, but existing routing methods often introduce additional training, deployment, and maintenance overhead. This raises a natural question: can a LoRA-based EPM bank be routed without maintaining an additional...","url_abs":"https://arxiv.org/abs/2607.04118","url_pdf":"https://arxiv.org/pdf/2607.04118v1","authors":"[\"Fengxian Ji\",\"Zhuohan Xie\",\"Jingpu Yang\",\"Fan Zhang\",\"Zirui Song\",\"Xiuying Chen\"]","published":"2026-07-05T05:11:24Z","proceeding":"cs.LG","tasks":"[\"cs.LG\",\"cs.AI\"]","methods":"[\"LoRA\",\"Generative Adversarial Network\"]","project_urls":"[\"https://anonymous.4open.science/r/Parametric-Memory-Decoding-872A/\"]","has_code":false}