HyoJung Han's Avatar

HyoJung Han

@hjhan.bsky.social

Ph.D. candidate @ UMD CS Clip lab | ex Intern @ Meta FAIR & Microsoft | Multilingual and Multimodal NLP. Machine Translation and Speech Translation. https://h-j-han.github.io/

281 Followers  |  669 Following  |  6 Posts  |  Joined: 20.09.2023  |  1.5532

Latest posts by hjhan.bsky.social on Bluesky

I’ll be presenting our work ✨VocADT, tomorrow at #ICLR2025
Check out our poster session: iclr.cc/virtual/2025...
πŸ—“οΈThu 24 Apr 3 p.m. - 5:30 p.m
πŸ“Hall 3 + Hall 2B #250
So excited to be attending @iclr_conf in Singapore πŸ‡ΈπŸ‡¬!

23.04.2025 10:02 β€” πŸ‘ 0    πŸ” 0    πŸ’¬ 0    πŸ“Œ 0
Preview
VocADT - a h-j-han Collection Adapters for Altering LLM Vocabularies: What Languages Benefit the Most? at ICLR2025

Check out our paper: arxiv.org/abs/2410.09644
Github: github.com/h-j-han/VocADT
Huggingface πŸ€— Models: huggingface.co/collections/...

22.01.2025 19:27 β€” πŸ‘ 3    πŸ” 0    πŸ’¬ 0    πŸ“Œ 0
Post image

Accepted at #ICLR2025✨

🧐Which languages benefit the most from vocabulary adaptation?

We introduce VocADT, a new vocabulary adaptation method with a vocabulary adapter.
We explore the impact of various adaptation strategies on languages with diverse scripts and fragmentation to answer this question

22.01.2025 19:27 β€” πŸ‘ 7    πŸ” 1    πŸ’¬ 2    πŸ“Œ 1

starter pack for the Computational Linguistics and Information Processing group at the University of Maryland - get all your NLP and data science here!

go.bsky.app/V9qWjEi

10.12.2024 17:14 β€” πŸ‘ 29    πŸ” 12    πŸ’¬ 1    πŸ“Œ 1

Awesome starter pack! Thank youπŸ‘

11.12.2024 16:45 β€” πŸ‘ 1    πŸ” 0    πŸ’¬ 0    πŸ“Œ 0

πŸ‘

05.12.2024 20:50 β€” πŸ‘ 1    πŸ” 0    πŸ’¬ 0    πŸ“Œ 0

Wow nice picture!

03.12.2024 18:10 β€” πŸ‘ 1    πŸ” 0    πŸ’¬ 1    πŸ“Œ 0
Preview
SpeechQE: Estimating the Quality of Direct Speech Translation HyoJung Han, Kevin Duh, Marine Carpuat. Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing. 2024.

Estimating translation quality is key to building trustworthy machine translation systems, but most work focuses on text. How well can we assess *speech* translation? 🎀
Check out our π’π©πžπžπœπ‘ππ„ πŸ’¬ paper with Hyojung Han and Kevin Duh at #EMNLP2024!
aclanthology.org/2024.emnlp-m...

12.11.2024 12:46 β€” πŸ‘ 6    πŸ” 1    πŸ’¬ 2    πŸ“Œ 0

@hjhan is following 20 prominent accounts