ProUIE: A Macro-to-Micro Progressive Learning Method for LLM-based Universal Information Extraction

2026-04-12

AI Summary

Existing large-model-driven general information extraction methods often rely on external resources, suffer from training complexity, and yield limited performance gains. This work proposes a macro–meso–micro progressive learning framework that requires no external data and enhances unified extraction capability through three stages: Complete Modeling (CM), Simplified Alignment (SA), and Deep Exploration (DE) powered by GRPO with Stepwise Fine-grained Rewards (SFR). The approach introduces, for the first time, a structure-unit-level fine-grained reward mechanism. Evaluated across 36 public datasets, it significantly outperforms strong baselines, achieving state-of-the-art average performance in named entity recognition and relation extraction with a smaller backbone model, and demonstrates tangible benefits in large-scale production environments.

Technology Category

• Natural Language Processing: Information Extraction
• Machine Learning: Large Multimodal Models (LMMs)
• Data Mining & Knowledge Management: Linked Open Data, Knowledge Graphs & KB Completion

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Web Mining and Content Analysis: Large pretrained models with web data

Abstract

LLM-based universal information extraction (UIE) methods often rely on additional information beyond the original training data, which increases training complexity yet often yields limited gains. To address this, we propose ProUIE, a Macro-to-Micro progressive learning approach that improves UIE without introducing any external information. ProUIE consists of three stages: (i) macro-level Complete Modeling (CM), which learns NER, RE, and EE along their intrinsic difficulty order on the full training data to build a unified extraction foundation, (ii) meso-level Streamlined Alignment (SA), which operates on sampled data with simplified target formats, streamlining and regularizing structured outputs to make them more concise and controllable, and (iii) micro-level Deep Exploration (DE), which applies GRPO with stepwise fine-grained rewards (SFR) over structural units to guide exploration and improve performance. Experiments on 36 public datasets show that ProUIE consistently improves unified extraction, outperforming strong instruction-tuned baselines on average for NER and RE while using a smaller backbone, and it further demonstrates clear gains in large-scale production-oriented information extraction.

Problem

• Universal Information Extraction
• LLM-based UIE
• Training Complexity
• External Information
• Unified Extraction

Innovation

• Progressive Learning
• Universal Information Extraction
• Macro-to-Micro
• GRPO
• Stepwise Fine-grained Rewards