arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2104.08799cs.CL
收藏

基于细粒度评估引导的强化学习关键词生成

Keyphrase Generation with Fine-Grained Evaluation-Guided Reinforcement Learning

Fudan University(复旦大学)

Yichao Luo, Yige Xu, Jiacheng Ye, Xipeng Qiu, Qi Zhang

摘要:

关键词生成(KG)是一个经典的任务,用于从给定的文档中捕获中心思想,以生成一组关键词。基于Seq2Seq模型,先前的KG任务上的强化学习框架利用评估指标来进一步改进训练良好的神经模型。然而,诸如$F_1@5$和$F_1@M$的这些KG评估度量仅知道短语级别上的预测的确切正确性,而忽略了相似预测和目标之间的语义相似性,这抑制了模型学习深度语言模式。针对这个问题,我们提出了一个新的细粒度的评价指标,以改善RL框架,它考虑不同的粒度:令牌级$F_1$分数,编辑距离,重复,和预测量。总体而言,新框架包括两个奖励函数:细粒度的评估分数和香草$F_1$分数。该框架有助于模型识别部分匹配短语,这些部分匹配短语可以进一步优化为精确匹配短语。KG基准测试的实验表明,我们提出的训练框架优于以前的RL训练框架的所有评价分数。此外,我们的方法可以有效地缓解同义词问题,并产生更高质量的预测。源代码可从\url{https://github.com/xuyige/FGRL4KG}获得。

英文摘要:

Aiming to generate a set of keyphrases, Keyphrase Generation (KG) is a classical task for capturing the central idea from a given document. Based on Seq2Seq models, the previous reinforcement learning framework on KG tasks utilizes the evaluation metrics to further improve the well-trained neural models. However, these KG evaluation metrics such as $F_1@5$ and $F_1@M$ are only aware of the exact correctness of predictions on phrase-level and ignore the semantic similarities between similar predictions and targets, which inhibits the model from learning deep linguistic patterns. In response to this problem, we propose a new fine-grained evaluation metric to improve the RL framework, which considers different granularities: token-level $F_1$ score, edit distance, duplication, and prediction quantities. On the whole, the new framework includes two reward functions: the fine-grained evaluation score and the vanilla $F_1$ score. This framework helps the model identifying some partial match phrases which can be further optimized as the exact match ones. Experiments on KG benchmarks show that our proposed training framework outperforms the previous RL training frameworks among all evaluation scores. In addition, our method can effectively ease the synonym problem and generate a higher quality prediction. The source code is available at \url{https://github.com/xuyige/FGRL4KG}.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑