arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2008.07688cs.LGcs.AIcs.IRstat.ML
收藏

基于自然语言推理的澄清问题排序

Ranking Clarification Questions via Natural Language Inference

Carnegie Mellon University(卡内基梅隆大学)

Vaibhav Kumar, Vikas Raunak, Jamie Callan

摘要:

给定一个自然语言查询,教机器提出澄清问题在实际的自然语言处理系统中具有巨大的效用。这种交互可以帮助填补信息空白,以便更好地理解查询。对于对澄清问题进行排名的任务,我们假设确定澄清问题是否与给定帖子(在StackExchange等QA论坛上)中的缺失条目有关可以被视为自然语言推理(NLI)的特殊情况,其中帖子和最相关的澄清问题都指向共享的潜在信息或上下文。我们通过将在NLI和多NLI数据集上微调的Siamese BERT模型的表示合并到我们的模型中来验证这一假设,并证明我们的最佳性能模型分别获得了40%和60%的相对性能提高(在关键指标Precision@1上),在StackExchange数据集的两个评估集的最新基线上,因此,远远超过了最先进的技术。

英文摘要:

Given a natural language query, teaching machines to ask clarifying questions is of immense utility in practical natural language processing systems. Such interactions could help in filling information gaps for better machine comprehension of the query. For the task of ranking clarification questions, we hypothesize that determining whether a clarification question pertains to a missing entry in a given post (on QA forums such as StackExchange) could be considered as a special case of Natural Language Inference (NLI), where both the post and the most relevant clarification question point to a shared latent piece of information or context. We validate this hypothesis by incorporating representations from a Siamese BERT model fine-tuned on NLI and Multi-NLI datasets into our models and demonstrate that our best performing model obtains a relative performance improvement of 40 percent and 60 percent respectively (on the key metric of Precision@1), over the state-of-the-art baseline(s) on the two evaluation sets of the StackExchange dataset, thereby, significantly surpassing the state-of-the-art.

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。

↑