3 ms·
Tel-Aviv University Releases Long-Text NLP Benchmark
- groffee 5y ago> We hope that SCROLLS inspires the NLP community to go beyond single sentences and paragraphs, and meet the challenges of processing and reasoning over longer discourses. Very passive aggressive.
- mistrial9 5y agoexcellent! irony here -- this single sentance can be taken as constructive and engaged OR a passive aggressive attack on current research and therefore researchers. No evidence in this sentance could prove either interpretation, nor others. So what is the article context? for that matter, the reputation of the speaker? What is the rest of the sentiment in this article text? What forum for publication? What is the typical audience sentiment of readers of this forum? all so important, right?
- jamra 5y agoHaving had experience with researchers at that university, I would be much more inclined to believe that their command over the English language is the culprit. I’ve helped write documents in English for one researcher there and everything reads as run on sentences. Please don’t judge non English speakers to the same standards.
- travisgriggs 5y agoDid anyone else glance at the title and get curious what the latest news was with the Dead Sea or other related ancient scrolls? If NLP had been spelled out, I might not have made the mistake. In retrospect, I know what NLP can mean, but I also know that just about every acronym has multiple meanings attached to it anymore.
- mistrial9 5y agohttps://en.wikipedia.org/wiki/NLP https://en.wikipedia.org/wiki/NLP
- WalterGR 5y ago> Did anyone else glance at the title and get curious Yes. Not about the "NLP" part, but about the improperly capitalized "Scrolls" part. Recently I've been seeing a lot of capitalization changes in HN titles vs. the original page titles. I believe HN automatically adjusts submitted titles in some ways - I wonder if this is one of them. Or if it's a bookmarklet. Or human error.