HTTP/2 200 server: GitHub.com content-type: application/xml last-modified: Tue, 23 Dec 2025 00:33:22 GMT access-control-allow-origin: * strict-transport-security: max-age=31556952 etag: W/"6949e352-7ab" expires: Mon, 29 Dec 2025 20:46:00 GMT cache-control: max-age=600 content-encoding: gzip x-proxy-cache: MISS x-github-request-id: EB36:1387E:93BE31:A5DC3C:6952E62F accept-ranges: bytes age: 0 date: Mon, 29 Dec 2025 20:36:00 GMT via: 1.1 varnish x-served-by: cache-bom-vanm7210080-BOM x-cache: MISS x-cache-hits: 0 x-timer: S1767040560.936743,VS0,VE223 vary: Accept-Encoding x-fastly-request-id: e89b0ff5442b5a40f11f1634e051bbbe822839c5 content-length: 762 Jekyll2025-12-22T16:33:07-08:00https://sophiahan6.github.io/feed.xmlpersonal descriptionSophia Hansophiahsm6@gmail.com2020-05-20T00:00:00-07:002020-05-20T00:00:00-07:00https://sophiahan6.github.io/posts/2020/05/blog-post-4Introduction

This post is for the paper [Resurrecting Submodularity in Neural Abstractive Summarization]

How to use submodularity in Text Generation?

Where is Coverage Related to?

Attention? Loss?

Submodular Optimization is impossible :(

Simple Workaround!

]]>Sophia Hansophiahsm6@gmail.com