Yufang Hou
@yufanghouProfessor for Natural Language Processing at IT:U - Interdisciplinary Transformation University Austria. Views are my own.
Similar User
@jessyjli
@thamar_solorio
@CopeNLU
@ReviewAcl
@HannaHajishirzi
@AnetteMFrank
@hhexiy
@swabhz
@aviaviavi__
@VioletNPeng
@royschwartzNLP
@boknilev
@vlachos_nlp
@dustin_wright37
@VeredShwartz
It was great talking about Holmes🔎 Poster improvisation is fun! #EMNLP2024 Great collaboration with @yotamperlitz @LChoshen @yufanghou @IGurevych More to come🔥 holmes-benchmark.github.io arxiv.org/abs/2404.18923
Prompting LMs is not enough to quantify their linguistic competence! Meet the Holmes🔎 benchmark at #EMNLP2024 #TACL or 👉🧵 💠Meta-study of current literature 💠Coverage of LMs and phenomena 💠Analysis of LM size, architecture, and instruction tuning holmes-benchmark.github.io
Meet @Furkansahinuc and @yufanghou now in Riverfront Hall and discover our LLM-based framework to create leaderboards aclanthology.org/2024.emnlp-mai…
Ever looked for paper leaderboards on certain tasks and datasets but find that community generated ones are inaccurate? 🥇🥈🥉 In our #EMNLP2024 paper, we introduce a dataset and an LLM-based framework to create accurate leaderboards! 🧵 (1/11) 📄 arxiv.org/abs/2409.12656
Woohoo 🥳Meet Holmes at #EMNLP2024 Interested in the linguistic competence of LMs and how it contrasts with their linguistic performance? Happy to chat and/or grab a drink!! Also, two of the amazing co-authors are around @yufanghou @LChoshen
Prompting LMs is not enough to quantify their linguistic competence! Meet the Holmes🔎 benchmark at #EMNLP2024 #TACL or 👉🧵 💠Meta-study of current literature 💠Coverage of LMs and phenomena 💠Analysis of LM size, architecture, and instruction tuning holmes-benchmark.github.io
And consider following the authors Furkan Şahinuç (@Furkansahinuc), Thy Thy Tran (@thy2512), Yulia Grishina, Yufang Hou (@yufanghou), Bei Chen, Iryna Gurevych (@IGurevych) for more information or an exchange of ideas. (10/🧵)
»Efficient Performance Tracking: Leveraging Large Language Models for Automated Construction of Scientific Leaderboards« by @Furkansahinuc, @thy2512, Yulia Grishina, @yufanghou, Bei Chen, and @IGurevych @UKPLab, @IBMResearch, @AmazonScience 📄 arxiv.org/abs/2409.12656 (3/🧵)
United States Trends
- 1. Cowboys 50,4 B posts
- 2. Bears 74,4 B posts
- 3. Chiefs 67,9 B posts
- 4. Panthers 42,1 B posts
- 5. Turpin 11,8 B posts
- 6. Texans 28,2 B posts
- 7. Commanders 56,7 B posts
- 8. Vikings 43,8 B posts
- 9. Caleb 34,5 B posts
- 10. Bryce Young 8.696 posts
- 11. #DALvsWAS 13,2 B posts
- 12. Eberflus 8.632 posts
- 13. Titans 46 B posts
- 14. #skol 5.555 posts
- 15. Jayden Daniels 7.665 posts
- 16. #KeepPounding 3.471 posts
- 17. Colts 29,9 B posts
- 18. Terry 32 B posts
- 19. CJ Stroud 4.419 posts
- 20. #RaiseHail 8.868 posts
Who to follow
-
Jessy Li
@jessyjli -
thamar |
@thamar_solorio -
CopeNLU
@CopeNLU -
ACLRollingReview
@ReviewAcl -
Hanna Hajishirzi
@HannaHajishirzi -
Anette Frank
@AnetteMFrank -
He He
@hhexiy -
Swabha Swayamdipta ✈️ EMNLP'24 🏖️
@swabhz -
Avi Sil
@aviaviavi__ -
Violet Peng
@VioletNPeng -
Roy Schwartz
@royschwartzNLP -
Yonatan Belinkov
@boknilev -
Andreas Vlachos
@vlachos_nlp -
Dustin Wright
@dustin_wright37 -
Vered Shwartz
@VeredShwartz
Something went wrong.
Something went wrong.