Too many papers to read? Try TLDR - Extreme Summarization of Scientific Documents

There are over 300 deep learning papers published every day. I find it very hard to keep up with. This paper introduces a cool way to summarize papers to extremely short summaries (TLDRs). More interestingly, Semantic Scholar uses the proposed method to build its TLDR feature and it's available in beta for nearly 10 million papers now!

0:00 - Too many papers
1:33 - What's special about this paper
2:25 - Sci-TLDR
5:16 - Controlled abstraction for TLDRs with title scaffolding
9:04 - During training
9:51 - Extractive summarization baselines
10:46 - Abstractive summarization baselines
11:14 - Input space
12:59 - Oracle
13:50 - ROGUE metrics
14:35 - Experiment results
17:25 - Model generated examples
17:59 - Demo - real-word application

Code and data:
TLDR Extreme Summarization of Scientific Documents
 https://github.com/allenai/scitldr

TLDR feature in Semantic Scholar: https://tldr.semanticscholar.org/

Abstract
We introduce TLDR generation, a new form of extreme summarization, for scientific papers. TLDR generation involves high source compression and requires expert background knowledge and understanding of complex domain-specific language. To facilitate study on this task, we introduce SciTLDR, a new multi-target dataset of 5.4K TLDRs over 3.2K papers. SciTLDR contains both author-written and expert-derived TLDRs, where the latter are collected using a novel annotation protocol that produces high-quality summaries while minimizing annotation burden. We propose CATTS, a simple yet effective learning strategy for generating TLDRs that exploits titles as an auxiliary training signal. CATTS improves upon strong baselines under both automated metrics and human evaluations.

Connect
Linkedin https://www.linkedin.com/in/xue-yong-fu-955723a6/
Twitter https://twitter.com/home
Email edwindeeplearning@gmail.com

Видео Too many papers to read? Try TLDR - Extreme Summarization of Scientific Documents канала Deep Learning Explainer