dorsal/arxiv
View SchemaNSR-Boost: A Neuro-Symbolic Residual Boosting Framework for Industrial Legacy Models
| Authors | Ziming Dai, Dabiao Ma, Jinle Tong, Mengyuan Han, Jian Yang, Haojun Fei |
|---|---|
| Categories | |
| ArXiv ID | 2601.10457vv1 |
| URL | https://arxiv.org/abs/2601.10457 |
| License | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ |
Abstract
Although the Gradient Boosted Decision Trees (GBDTs) dominate industrial tabular applications, upgrading legacy models in high-concurrency production environments still faces prohibitive retraining costs and systemic risks. To address this problem, we present NSR-Boost, a neuro-symbolic residual boosting framework designed specifically for industrial scenarios. Its core advantage lies in being "non-intrusive". It treats the legacy model as a frozen model and performs targeted repairs on "hard regions" where predictions fail. The framework comprises three key stages: first, finding hard regions through residuals, then generating interpretable experts by generating symbolic code structures using Large Language Model (LLM) and fine-tuning parameters using Bayesian optimization, and finally dynamically integrating experts with legacy model output through a lightweight aggregator. We report on the successful deployment of NSR-Boost within the core financial risk control system at Qfin Holdings. This framework not only significantly outperforms state-of-the-art (SOTA) baselines across six public datasets and one private dataset, more importantly, shows excellent performance gains on real-world online data. In conclusion, it effectively captures long-tail risks missed by traditional models and offers a safe, low-cost evolutionary paradigm for industry.
{
"annotation_id": "858947bb-3211-4110-ae70-1578c023f453",
"date_created": "2026-02-17T05:53:23.417000Z",
"date_modified": "2026-02-17T05:53:23.417000Z",
"file_hash": "a05646917ed5bd11a8ee0c3dfe956bbb32b1d8caab9a19aa4c0c5c2c3f6fb2d7",
"private": false,
"record": {
"abstract": "Although the Gradient Boosted Decision Trees (GBDTs) dominate industrial tabular applications, upgrading legacy models in high-concurrency production environments still faces prohibitive retraining costs and systemic risks. To address this problem, we present NSR-Boost, a neuro-symbolic residual boosting framework designed specifically for industrial scenarios. Its core advantage lies in being \"non-intrusive\". It treats the legacy model as a frozen model and performs targeted repairs on \"hard regions\" where predictions fail. The framework comprises three key stages: first, finding hard regions through residuals, then generating interpretable experts by generating symbolic code structures using Large Language Model (LLM) and fine-tuning parameters using Bayesian optimization, and finally dynamically integrating experts with legacy model output through a lightweight aggregator. We report on the successful deployment of NSR-Boost within the core financial risk control system at Qfin Holdings. This framework not only significantly outperforms state-of-the-art (SOTA) baselines across six public datasets and one private dataset, more importantly, shows excellent performance gains on real-world online data. In conclusion, it effectively captures long-tail risks missed by traditional models and offers a safe, low-cost evolutionary paradigm for industry.",
"arxiv_id": "2601.10457",
"authors": [
"Ziming Dai",
"Dabiao Ma",
"Jinle Tong",
"Mengyuan Han",
"Jian Yang",
"Haojun Fei"
],
"categories": [
"cs.AI"
],
"license": "http://arxiv.org/licenses/nonexclusive-distrib/1.0/",
"title": "NSR-Boost: A Neuro-Symbolic Residual Boosting Framework for Industrial Legacy Models",
"url": "https://arxiv.org/abs/2601.10457",
"version": "v1"
},
"schema_id": "dorsal/arxiv",
"source": {
"execution_id": "f3b44656-d948-4464-9ab5-31a6177f497b",
"id": "arXiv Dataset",
"type": "Model",
"variant": "snapshot-2026-01-17",
"version": "0.1.0"
},
"user_id": 1000002
}