关于词汇替换的学术论文从arXiv撤回

作者 PulseAugur 编辑部 · [1 个来源] · 2026-11-14 05:00

一篇题为“Unsupervised Candidate Ranking for Lexical Substitution via Holistic Sentence Semantics”的研究论文被提交至arXiv，提出了用于词汇替换任务中候选词排序的新方法。该论文由Zhongyang Hu撰写，研究了使用注意力权重和集成梯度来衡量上下文影响和句子相似度的方法。在LS07和SWORDS数据集上的实验显示了排序性能的提高，但该论文后来已被撤回。 AI

影响这项关于词汇替换的研究，虽然已被撤回，但探索了改进语言模型语义理解的新方法。

排序理由该集群描述了一篇关于特定NLP任务的已撤回的学术论文。[lever_c_demoted from research: ic=1 ai=1.0]

在 arXiv cs.AI 阅读 →

AI 生成摘要 · Google Gemini · 来自 1 个来源。我们如何撰写摘要 →

本文如何被排名

Signal score: 16 / 100; Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket: Tool; 该集群描述了一篇关于特定NLP任务的已撤回的学术论文。[lever_c_demoted from research: ic=1 ai=1.0]
Source corroboration: Single-source cluster; Only one publisher covered this so far. Single-source stories can still rank when the publisher is high-authority, but they lack cross-source corroboration.
Topics: paper, other; Editorial topic classification. Feeds into how the story surfaces on /topic/<slug> hub pages and into the per-entity coverage mix.
AI-industry relevance: High; Clearly on-topic for AI-industry coverage.
Story freshness: Breaking (< 6h); Fresh story with cross-source coverage still developing. Ranking may shift as more sources report.

完整方法见我们的编辑标准。

报道来源 [1]

arXiv cs.AI TIER_1 English(EN) · Zhongyang Hu, Naijie Gu, Xiangzhi Tao, Tianhui Gu, Yibing Zhou · 2026-11-14 05:00

基于整体句子语义的词汇替换无监督候选词排序

arXiv:2509.11513v2 Announce Type: replace-cross Abstract: A key subtask in lexical substitution is ranking the given candidate words. A common approach is to replace the target word with a candidate in the original sentence and feed the modified sentence into a model to capture s…

本文如何被排名

报道来源 [1]

基于整体句子语义的词汇替换无监督候选词排序

相关实体

相关话题