NSR-Boost: A Neuro-Symbolic Residual Boosting Framework for Industrial Legacy Models

TL;DR

NSR-Boost is a neuro-symbolic framework that enhances legacy industrial models by targeted, interpretable repairs, significantly improving performance while minimizing retraining costs and systemic risks.

Contribution

It introduces a non-intrusive, residual boosting approach using LLM-generated symbolic experts for industrial model repair and enhancement.

Findings

Outperforms state-of-the-art baselines on multiple datasets

Successfully deployed in a real-world financial risk system

Reduces bad rate and captures long-tail risks effectively

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. Experimental results demonstrate that the framework not only significantly outperforms state-of-the-art (SOTA) baselines across six public datasets and one private dataset. More importantly, we report the successful deployment of NSR-Boost within the core financial risk control system of Qfin Holdings, where empirical results on real-world online traffic exhibit superior performance improvements and a significant reduction in the bad rate. In conclusion, it effectively captures long-tail risks missed by traditional models and offers a safe, low-cost evolutionary paradigm for industry.