MR-UIE: Multi-Perspective Reasoning with Reinforcement Learning for Universal Information Extraction
This work addresses a domain-specific problem for researchers and practitioners in information extraction, offering an incremental improvement by enhancing generalization in complex tasks.
The paper tackles the problem of insufficient performance of large language models in universal information extraction, especially for structured outputs requiring multi-step reasoning, by integrating reinforcement learning with multi-perspective reasoning, resulting in improved extraction accuracy across domains and surpassing state-of-the-art methods on several datasets.
Large language models (LLMs) demonstrate robust capabilities across diverse research domains. However, their performance in universal information extraction (UIE) remains insufficient, especially when tackling structured output scenarios that involve complex schema descriptions and require multi-step reasoning. While existing approaches enhance the performance of LLMs through in-context learning and instruction tuning, significant limitations nonetheless persist. To enhance the model's generalization ability, we propose integrating reinforcement learning (RL) with multi-perspective reasoning for information extraction (IE) tasks. Our work transitions LLMs from passive extractors to active reasoners, enabling them to understand not only what to extract but also how to reason. Experiments conducted on multiple IE benchmarks demonstrate that MR-UIE consistently elevates extraction accuracy across domains and surpasses state-of-the-art methods on several datasets. Furthermore, incorporating multi-perspective reasoning into RL notably enhances generalization in complex IE tasks, underscoring the critical role of reasoning in challenging scenarios.