Butoi Alexandra (@butoialexandra) 's Twitter Profile
Butoi Alexandra

@butoialexandra

PhD student at ETH Zürich focusing on NLP

ID: 784163553230069760

calendar_today06-10-2016 22:48:54

16 Tweet

144 Followers

209 Following

Vilém Zouhar (@zouharvi) 's Twitter Profile Photo

Nothing beats Goldsmith 2001: > writes a paper > makes a decision > "without much conviction behind decision" > refuses to elaborate > end of section Butoi Alexandra @ryandcotterell aclanthology.org/J01-2001.pdf

Nothing beats Goldsmith 2001:
> writes a paper
> makes a decision
> "without much conviction behind decision"
> refuses to elaborate
> end of section
<a href="/butoialexandra/">Butoi Alexandra</a> @ryandcotterell aclanthology.org/J01-2001.pdf
Niklas Stoehr (@niklas_stoehr) 's Twitter Profile Photo

Long queues to get a chance to speak with Anej Svete and Shauli Ravfogel! 🐍 Check out their work on Acyclic WFSA with Failure Arcs and Kernelized Concept Erasure EMNLP 2025! @ryandcotterell Jason Eisner Tim Vieira Benjamin Dayan ETH CS Department cs.jhu.edu/~jason/papers/… arxiv.org/abs/2201.12191

Long queues to get a chance to speak  with <a href="/AnejSvete/">Anej Svete</a> and <a href="/ravfogel/">Shauli Ravfogel</a>! 🐍 Check out their work on Acyclic WFSA with Failure Arcs and Kernelized Concept Erasure <a href="/emnlpmeeting/">EMNLP 2025</a>! @ryandcotterell <a href="/adveisner/">Jason Eisner</a> <a href="/xtimv/">Tim Vieira</a> <a href="/BenjaminDayan/">Benjamin Dayan</a> <a href="/CSatETH/">ETH CS Department</a>

cs.jhu.edu/~jason/papers/…

arxiv.org/abs/2201.12191
Franz Nowak (@franz_nowak) 's Twitter Profile Photo

Excited to be presenting my first paper, co-authored with @ryandcotterell, at ACL 2023 in Toronto this July! We developed a new fast prefix parsing algorithm that can be used for language models. arxiv.org/abs/2306.02303

David Chiang (@davidweichiang) 's Twitter Profile Photo

To appear at ACL, with Alexandra Butoi Butoi Alexandra and @ryandcotterell: we make precise in what way a tree-adjoining grammar is like a CFG in a CFG, a linear indexed grammar is like a PDA in a CFG, and an embedded PDA is like a PDA in a PDA arxiv.org/abs/2306.03628

To appear at ACL, with Alexandra Butoi <a href="/butoialexandra/">Butoi Alexandra</a> and @ryandcotterell: we make precise in what way a tree-adjoining grammar is like a CFG in a CFG, a linear indexed grammar is like a PDA in a CFG, and an embedded PDA is like a PDA in a PDA 
arxiv.org/abs/2306.03628
ACL 2025 (@aclmeeting) 's Twitter Profile Photo

ACL org announcement: 📢 The list of accepted tutorials at EACL, ACL, NAACL and EMNLP in 2024 is out 🎉 #NLProc Please help spread the word. Retweeting w/ references, especially with instructors information etc is very much appreciated - thanks! 😊

David Chiang (@davidweichiang) 's Twitter Profile Photo

Alexandra Butoi Butoi Alexandra, Tim Vieira Tim Vieira, @ryandcotterell, and David Chiang. Efficient algorithms for recognizing weighted tree-adjoining languages. A unified framework for parsing with TAG, LIG, EPDA, and PAAs, with best-ever asymptotic complexities in some cases.

David Chiang (@davidweichiang) 's Twitter Profile Photo

Butoi Alexandra will present her #EMNLP2023 paper on parsing weighted tree-adjoining languages at 9am in the East Foyer! virtual2023.emnlp.org/paper_4629.html arxiv.org/abs/2210.06884

Afra Amini (@afra_amini) 's Twitter Profile Photo

Current KL estimation practices in RLHF can generate high variance and even negative values! We propose a provably better estimator that only takes a few lines of code to implement.🧵👇 w/ Tim Vieira and Ryan Cotterell code: arxiv.org/pdf/2504.10637 paper: github.com/rycolab/kl-rb

Current KL estimation practices in RLHF can generate high variance and even negative values! We propose a provably better estimator that only takes a few lines of code to implement.🧵👇
w/ <a href="/xtimv/">Tim Vieira</a> and Ryan Cotterell
code: arxiv.org/pdf/2504.10637
paper: github.com/rycolab/kl-rb