论文修改建议 (ZhangYX 20211109 摘要第一句)

1. 第一句不够亲民. 没有给读者机器学习领域的感觉. 随着什么领域/应用的发展, 促进了set-level问题的发展. 以下是修改的各个版本, 可以体会句子的锤炼过程. 以最终版为准.
   Set-level problems becomes more important with the advance of complex machine learning problems on imges, texts, medicine.
   Set-level problems becomes critical with the advance of complex machine learning problems on imges, texts, medicine.
   Set-level problems becomes critical with the complication of machine learning problems on imges, texts, medicine.
   Set-level problems becomes critical with the rapid advance of machine learning to real world applications concerning images, texts, and medicine.
   Set-level problems becomes critical when we are interested in animals in pictures, people in novels, and componts in medicine. (这个是最终版)

with the rapid advance of machine learning to real world applications concerning images, texts, and medicine.

Set-level models aim to handle animals in pictures, links in web pages, and componts in drugs.

Set-level problems becomes critical when we are interested in animals in pictures, links in web pages, and componts in drugs.
The key issue is to measure the similarity between two sets.

从小到大，从实际到理论。说得具体，至于推广是读者的事情。

2. MSK indirectly measures the similarity between two sets by determining the relationship between their embedding vectors.

MSK indirectly measures set similarity by determining the relationship between embedding vectors.