NE-R1：通过强化学习增强命名实体识别模型
NE-R1: Enhancing Named Entity Recognition Model via Reinforcement Learning
另 1 家 · 查看机构详情
- Shenzhen Graduate School, Peking University（北京大学深圳研究生院）
- Baidu, Inc.（百度公司）
- Shenzhen University（深圳大学）
- Gaoling School of Artificial Intelligence, Renmin University of China（中国人民大学高瓴人工智能学院）
机构由 AI 辅助整理，请以论文原文为准。
浏览论文内容
中文总结 AI 辅助
该研究针对NER中长尾与领域实体识别难题，提出NE-R1框架，通过按需检索机制与两阶段训练结合强化学习优化，在多基准上实现F1分数提升，达SOTA性能。
中文摘要 AI 辅助
自大型语言模型（LLMs）问世以来，命名实体识别（NER）已取得显著进展。然而，由于参数化知识不足，长尾实体和领域特定实体的识别仍具挑战性。检索增强生成（RAG）提供了一种有前景的解决方案，可注入外部知识，但在处理熟悉案例时会引入噪声并产生不必要的成本。本文提出一种用于自适应检索增强NER的新型框架NE-R1，设计了NER的“按需检索”机制，并通过两阶段训练方法将其集成到模型中：（1）多任务指令调优初始化；（2）结合思维链（CoT）的端到端强化学习（RL）优化。为实现参数化知识与外部知识间的合理选择，设计了兼顾准确率与检索效益的多维奖励。NE-R1在多个基准上取得了最先进的性能，在领域内评估中平均F1分数提升2.52%，在零样本跨领域评估中平均F1分数提升1.18%。
英文摘要
Named Entity Recognition (NER) has achieved substantial progress since the advent of large language models (LLMs). Nevertheless, the recognition of long-tail and domain-specific entities remains challenging due to the deficiency in parametric knowledge. Retrieval-augmented generation (RAG) offers a promising remedy by injecting external knowledge, but it also introduces noise and unnecessary cost when dealing with familiar cases. In this paper, we propose NE-R1, a novel framework for adaptive retrieval-augmented NER. We design a "retrieval-on-demand" mechanism for NER. Then we integrate it into models by a two-stage training method: (1) multi-task instruction tuning initialization; (2) end-to-end RL optimization with CoT. To achieve reasonable selection between parameterized and external knowledge, we design a multi-dimensional reward considering both accuracy and retrieval benefit. NE-R1 achieves state-of-the-art performance on various benchmarks, with an average F1 score gain of 2.52% in in-domain evaluation and 1.18% in zero-shot cross-domain evaluation.
