English(EN) The Long Tail, Not the Front Page: Cold-Start Prediction of Crowd Highlight Salience

AI模型预测文档中的读者高亮内容

作者 PulseAugur 编辑部 · [3 个来源] · 2026-06-10 04:37

研究人员开发了一个模型，能够预测文档中的哪些段落会被读者高亮，即使在高亮内容尚未积累之前。该模型基于现有高亮数据进行训练，其表现比简单的基于开头的基线模型有微小但统计学上显著的优势。该系统在不太受欢迎的内容方面尤其有前景，其预测准确性在这些内容上更为突出。 AI

影响这项研究可以通过预测用户对特定段落的兴趣来改进内容摘要和推荐系统。

排序理由该集群包含一篇详细介绍新AI模型及其评估的学术论文。

AI 生成摘要 · Google Gemini · 来自 3 个来源。我们如何撰写摘要 →

报道来源 [3]

arXiv cs.CL TIER_1 English(EN) · Kazuki Nakayashiki, Keisuke Watanabe · 2026-06-11 04:00

The Long Tail, Not the Front Page: Cold-Start Prediction of Crowd Highlight Salience

arXiv:2606.11654v1 Announce Type: cross Abstract: A social highlighter's most useful signal -- which passages a crowd of readers marks -- exists only for documents people have already read. Can the aggregate crowd salience of a document be predicted from its text before its marks…
arXiv cs.IR (Information Retrieval) TIER_1 English(EN) · Keisuke Watanabe · 2026-06-10 04:37

长尾效应而非头条：冷启动预测人群高亮显著性

A social highlighter's most useful signal -- which passages a crowd of readers marks -- exists only for documents people have already read. Can the aggregate crowd salience of a document be predicted from its text before its marks accumulate? Prior work on this data found that ze…
arXiv cs.IR (Information Retrieval) TIER_1 English(EN) · Keisuke Watanabe · 2026-06-10 04:37

The Long Tail, Not the Front Page: Cold-Start Prediction of Crowd Highlight Salience

A social highlighter's most useful signal -- which passages a crowd of readers marks -- exists only for documents people have already read. Can the aggregate crowd salience of a document be predicted from its text before its marks accumulate? Prior work on this data found that ze…