Seth Ebner

Research scientist @Kensho. PhD from @jhuclsp. #NLProc

[Seth Ebner @_t_sne]

Come talk to us Tuesday, July 7 @ 11:00-12:30 PT in the Coronado room (Poster Session G)!

[Seth Ebner @_t_sne]

Kensho will present “On Finding Inconsistencies in Documents” and its dataset FIND at ACL! We show how well models find inconsistencies in complex documents, and we brought the receipts! Come by our poster July 7 @ 11:00-12:30 PT to learn more and get your own document debugged!

[Seth Ebner @_t_sne]

Dataset: https://t.co/ibavroXBAv

[Zhengping Jiang @zhengping_jiang]

New paper🚀! Factual precision metrics like FActScore are vulnerable to obvious or repetitive claims. We address this by introducing the 𝑪𝒐𝒓𝒆 process: a plug-and-play component that improves popular factual precision metrics through unique and informative subclaim selection.

[Shijie Wu @EzraWu]

Q: What's the best choice to improve zero-shot cross-lingual transfer performance if u can’t manually annotate any data?
A: The best setup is task dependent, so try multiple setups! 📈

Check out our #EMNLP2021 paper “Everything Is All It Takes” https://t.co/Ve11uxgFIC [1/4]

[JHU CLSP @jhuclsp]

"Everything Is All It Takes: A Multipronged Strategy for Zero-Shot Cross-Lingual Information Extraction"

Augmenting source language data with target language silver data -- created via MT, word alignment, annotation projection -- improves zero-shot cross-lingual IE. #emnlp2021

[Elias Stengel-Eskin @EliasEskin]

Very excited to share our new TACL paper, "Joint Universal Syntactic and Semantic Parsing", with @kentonmurray, @sheng_zh, @aaronsteven, and @ben_vandurme!

Paper: https://t.co/H25vVncxpn
Code: https://t.co/MpIkTvdaRW

🧵 1/14

[Kenton Murray @kentonmurray]

#NLPProc we are hiring summer interns for Cross-Lingual NLP research tasks. Apply here: https://t.co/BHd8RaOA7P

[Kenton Murray @kentonmurray]

“Gradual Fine-Tuning for Low-Resource Domain Adaptation” was just accepted at Adapt-NLP. It was great working with @_t_sne @aaronsteven @ben_vandurme and Mahsa Yarmohammadi. But I want to give shoutout to 1st author @fe1ixxu is applying to PhD programs right now.

[Seth Ebner @_t_sne]

New dataset—RAMS 🐏—for events w/ arguments across sentences! Larger + broader coverage than prior datasets! Plus document-level model for the task. “Multi-Sentence Argument Linking” w/ Patrick Xia, Ryan Culkin, @icoson, @ben_vandurme @jhuclsp #acl2020 Q+A July 8 17:00+20:00 GMT

[Arya McCarthy @ACL2025 🇦🇹 @aryamccarthy]

If your paper is arXived but you don't tweet it…

Proved a No Free Lunch theorem 🚫🍱 for clustering tasks, using our recently advocated metric. No best algo! With @ctongfei and @_t_sne, who survived my quibbles about \mathcal vs \mathbb \vs \mathfrak!

https://t.co/NA3uabgsKN