Omer Shubi retweeted
Imagine you are reading this post with some specific question in mind.โ๐ For example, is this yet another person who is going to brag about their ICLR paper acceptance? It turns out that it's possible decode which question you had in mind from your eye movements! ๐ ๐คฏ
Omer Shubi retweeted
Now out in @ScientificData, OneStop a 360-Participant English Eye Tracking Dataset with Different Reading Regimes.
Check it out here: nature.com/articles/s41597-0โฆ
Omer Shubi retweeted
๐ Fascinated by how we read and process language? Explore how eye movement while reading can be used in NLP. Join #ACL2025NLP with the "Eyetracking and NLP" tutorial! How eyetracking inform NLP and how NLP can contribute to modeling.
#Eyetracking #Psycholinguistics #NLProc
Omer Shubi retweeted
๐ Fascinated by how we read and process language? Explore how eye movement while reading can be used in NLP. Join #ACL2025NLP with the "Eyetracking and NLP" tutorial!
#Eyetracking #Psycholinguistics #NLProc
2025.aclweb.org/program/tutoโฆ
Omer Shubi retweeted
๐๐Big news! ๐๐
Happy to announce the release OneStop Eye Movements!๐พ๐พ
The OneStop dataset is the product of over 6 years of experimental design, data collection and data curation.
github.com/lacclab/OneStop-Eโฆ
Omer Shubi retweeted
๐ Our paper "๐๐๐ฏ๐ข๐ ๐๐ญ๐ข๐ง๐ ๐๐ฎ๐ฅ๐ญ๐ฎ๐ซ๐๐ฅ ๐๐ก๐๐ฌ๐ฆ๐ฌ: ๐๐ฑ๐ฉ๐ฅ๐จ๐ซ๐ข๐ง๐ ๐๐ง๐ ๐๐ง๐ฅ๐จ๐๐ค๐ข๐ง๐ ๐ญ๐ก๐ ๐๐ฎ๐ฅ๐ญ๐ฎ๐ซ๐๐ฅ ๐๐๐ ๐จ๐ ๐๐๐ฑ๐ญ-๐ญ๐จ-๐๐ฆ๐๐ ๐ ๐๐จ๐๐๐ฅ๐ฌ" is accepted to ๐๐๐๐!
@bd_eyal @roireichart @annalkorhonen
arxiv.org/abs/2310.01929
1/11
Omer Shubi retweeted
Excited to share our CoNLL 2024 paper on two open frontiers for psycholinguistics and language modeling: (1) goal driven comprehension (i.e. information seeking) and (2) repeated processing, with ๐๐ data.
Work by outstanding undergrad @Keren_G_K, @YoavMeiri and @scaperex.
Omer Shubi retweeted
Is it possible to predict human reading comprehension solely form eye movements during reading? Check out our lab's EMNLP 2024 paper by @scaperex, @YoavMeiri and Cfir Hadar, that studies this long-standing question systematically.
TLDR: it's very hard! arxiv.org/abs/2410.04484









