3 ms·
What you get from applying TLDR to their paper: We introduce SCITLDR, a new multi-target data set of 5.4KTLDRs over 3.2Kpapers. Keeping pdf's copy-paste artif
by justarandomq 6y ago
What you get from applying TLDR to their paper:
We introduce SCITLDR, a new multi-target data set of 5.4KTLDRs over 3.2Kpapers.
Keeping pdf's copy-paste artifacts:
We introduceTLDRgeneration, a new formof extreme extreme summarization, for scientific pa-pers.
Adding intro and conclusion (optional):
We introduce SCITLDR, a new data set of 5.4KTLDRs over 3.2Kpapers.
[0] https://scitldr.apps.allenai.org/ https://scitldr.apps.allenai.org/
- m1sta_ 6y agoYikes
- austinjp 6y agoYeah. I happen to have been looking at this problem in my spare time recently. I tried a bunch of abstractive AIs and approaches, and none produce consistently usable results. I'm sticking with extractive approaches plus a bunch of hard-coded general and domain-specific rules for now.