maieutic

This is the official account of the Maieutic Lab at JHU. We broadly work on Multilingual NLP and AI. The photos are what OpenAI's DALL-E "thinks" we are...

[Kenton Murray @kentonmurray]

Looking forward to visiting UVA this Friday. If you are around Charlottesville, I’d love to chat more about multilingual challenges in AI.

https://t.co/HLDvR79ECJ

[maieutic @maieuticlab]

Congrats to MAIEUTIC Lab’s first PhD student Haoran Xu @fe1ixxu who just won Honorable Mention in the ACL Dissertation Award! @aclmeeting

[maieutic @maieuticlab]

The MAIEUTIC Lab (https://t.co/ejVaEe5e26) is @aclmeeting and looking for new members. If you are interested in joining, please fill out this form: https://t.co/vGkUhp6oL4 If you are in person at ACL 2026, please fill this one out as well: https://t.co/fJ8xdbF1PZ

[MAGMaR @MAGMaR_workshop]

Dr. Nanyun (Violet) Peng is giving a keynote "Towards Self-Improving Multimodal Models" @VioletNPeng

[MAGMaR @MAGMaR_workshop]

Happening Now! MAGMaR 2026 @aclmeeting in the Old Town Room. Come chat about Multimodal Retrieval and Generation.

[maieutic @maieuticlab]

Come check out Neha’s work!

[Neha Verma @n_verma1]

I'll be in Seoul for #ICML2026 🇰🇷!

I'll present "DOTResize: Reducing LLM Width via Discrete Optimal Transport-based Neuron Merging" at the WeightSym workshop

I'm interested in LLM efficiency, KV cache, management, continual learning,... and birding 🐦. DM if you want to chat!

[maieutic @maieuticlab]

Can you generate grounded reports using videos instead of text documents for your LLM? Check out MAGMaR 2026 @aclmeeting this week!

[MAGMaR @MAGMaR_workshop]

🧵 The MAGMaR 2026 Shared Task results are out!

We challenged teams to retrieve relevant videos from a corpus of 110K+ multilingual clips AND generate grounded, persona-driven articles from them.

Come see the findings at our workshop at ACL on July 4!

https://t.co/9YNvHbEYKH

[Kenton Murray @kentonmurray]

Excited for MAGMaR 2026 in two days. People worry about LLMs hallucinating, and often you use RAG to help mitigate that. But what if your evidence isn’t in text? We ran a whole shared task (and workshop) where you use multimodal sources instead of documents - including videos.

[maieutic @maieuticlab]

The MAIEUTIC Lab is moving to GMU. We look forward to growing in size in the coming years!

[Kenton Murray @kentonmurray]

I'm excited to announce that this Fall I will be joining the Computer Science Department at George Mason University as an Asst. Prof. I'll be expanding my lab and looking for PhD students to work on Multilingual AI problems text, video, and speech.

https://t.co/f6PIESWRp4

[Neha Verma @n_verma1]

Introducing "DOTResize: Reducing LLM Width via Discrete Optimal Transport-based Neuron Merging" !

We introduce an optimal transport framework for Transformer width compression that redistributes signal across neurons rather than eliminating them 🚚⚖️

🧵 1/6

[maieutic @maieuticlab]

JHU community: Come see what my students built this semester.

[JHU Computer Science @JHUCompSci]

Take a break from studying and test your Wordle skills with a Wheel of Fortune! tournament tomorrow from 6–8 p.m. in Room 213 of the Bloomberg Student Center! Sign up now to play against other students and the AI bots they developed over the fall semester: https://t.co/nD0vUgG3Lp

[Kenton Murray @kentonmurray]

Super excited for the 1st Workshop on Multilingual Data Quality Signals (WMDQS) which is happening at #COLM2025 tomorrow. We are focused on looking all the way back to the web data that goes into all your LLMs and how we can do better at multilingual. Stop by! @COLM_conf

[maieutic @maieuticlab]

Check out our new lab website! https://t.co/ejVaEe4Gcy We are always looking for new collaborations - so reach out if you like one of our projects.

[Kenton Murray @kentonmurray]

Congrats to Dr. Xu @fe1ixxu on a successful thesis defense of "Minimizing Language Interference for Multilingual Models". The thesis covered 9 of his first author papers encompassing:

Modeling (https://t.co/VWXCvZyUrG, https://t.co/pOIq2jpoeg, https://t.co/9HXuUhBXXz)

[Kenton Murray @kentonmurray]

Introducing X-ALMA: a 50 language multilingual machine translation model. It’s average translation performance outperforms other multilingual models (including ones focused on fewer num of langs) pushing against the curse of multilinguality.

[Niyati Bafna @BafnaNiyati]

📢When LLMs solve tasks with a mid-to-low resource input/target language, their output quality is poor. We know that. But can we pin down what breaks inside the LLM? We introduce the 💥translation barrier hypothesis💥 for failed multilingual generation. https://t.co/VnrOWdNPr8

[maieutic @maieuticlab]

A nice summary of some of our work at NAACL earlier this year for a non-technical audience.

[JHU Computer Science @JHUCompSci]

A new study from Johns Hopkins researchers @nikhilsksharma, @ZiangXiao, and @kentonmurray finds that multilingual #AI privileges dominant languages, deepening divides rather than democratizing access to information. Read more: https://t.co/Z8VAPrR3Xq

[Kenton Murray @kentonmurray]

عسلامة,

As part of IWSLT 2023, we are hosting a Tunisian-to-English Speech Translation Shared Task and would love people to participate. The evaluation campaign runs April 1st-15th.

More details can be found here: https://t.co/0GbDB8cNFr

Yaishek

[Stella Li @StellaLisy]

How to fine-tune a model when you have close to no labeled data? Check out our new paper "Language Agnostic Code-Mixing Data Augmentation by Predicting Linguistic Patterns" which introduces a zero-cost code-mixing generation method for sentiment analysis: https://t.co/FC9goOJcRI

[Neha Verma @n_verma1]

Check out our #EMNLP2022 paper on improving the isomorphism of word embedding spaces for bilingual lexicon induction! Led by the awesome @cheeesio