NLPLab@Montclair

@LinguisticsMSU Congratulations! @montclairstateu @LinguisticsMSU

Congratulations, Sara Flores and Dr. Covey ! Really nice piece: montclair.edu/newscenter/202… It’s great to see Sara’s effort and enthusiasm come through, and to see Dr. Covey’s impact in the story, too!

Oct 27 at Montclair: @zawadzkanatalka on how European media racializes and genders refugee crises—same violence, different headlines. youtube.com/watch?v=riObdP… @Montclair_CHSS #montclair #linguistics @lab_nlp

Looking for scholars whose work addresses real-world language challenges in educational, clinical, or communicative contexts, grounded in strong theoretical foundations.

Takeaway: cross-lingual transfer isn’t “more is better.” Success depends on the model, language pair, and training order. Great work by Julia Sammartino advancing research on multilingual euphemism detection! 👏 #RANLP2025 #TransferLearning #LowResource #AIResearch

🧩 Findings: • XLM-R → bigger boosts, but more catastrophic forgetting when languages differ. • mBERT → steadier, smaller gains, less forgetting. Transfer helps, but only in the right conditions.

🎉 We’re proud of our recent graduate Julia Sammartino linkedin.com/in/julia-samma…, who presented her paper “When Does Language Transfer Help? Sequential Fine-Tuning for Cross-Lingual Euphemism Detection” at #RANLP2025! 📄 Preprint: arxiv.org/abs/2508.11831

Linguistics kicked off its new Book Club with Algospeak by @adam_aleksic how people twist words like “unalive” or “seggs” to dodge algorithms. Takeaway: platforms don’t just host language -- they shape it. Next up: Cultish by @amandamontell ! #MontclairLinguistics #Algospeak

See this Instagram post by @montclair_linguistics instagram.com/p/DPzzh83CYWE/…

Dr. Suma Bhat at Montclair! @Montclair_CHSS @LinguisticsMSU @MSUCSAM

New Talk on YouTube! "Discovering Implicit Social Biases in Large Language Models" by @sharonlevy21 . Explore how hidden biases in AI affect decisions in healthcare, education, and more! youtu.be/-weamEyPiQY #NLPMontclair #GenderBias #FairAI @LinguisticsMSU @montclairstateu

🌟 Excellent presentations today at the Graduate Mini-Conference Fall 2024! 🎓👏 Huge congratulations to all the students who shared their work. Excited to see your future endeavors.🚀 #GradConference2024 #ResearchExcellence @Montclair_CHSS @MSUCSAM @lab_nlp @LibbyBarak

Today!

Join us Nov 21, 2024, at 3:30 PM for a talk by Dr. Sharon Levy @sharonlevy21: "Discovering Implicit Social Biases in Large Language Models". Explore how AI biases arise unexpectedly in domains like healthcare, education, and decision-making. #NLPMontclair #ResponsibleAI

Great to see both new and returning students collaborating at the NLP lab! Exciting discussions ahead.@LibbyBarak @FeldmanMSU @Montclair_CHSS @MSUCSAM @montclairstateu @LinguisticsMSU #MontclairNLP

Our own Dr. Aparna Varde and the Prime Minister of Lithuania, Ingrida Šimonytė @IngridaSimonyte at the ECML-PKDD @ECMLPKDD conference! @MSUCSAM @Montclair_CHSS #MontclairNLP

Our lab's new paper, "Turkish Delights: a Dataset on Turkish Euphemisms," offers a new annotated dataset of euphemisms in Turkish. Read the paper here: arxiv.org/abs/2407.13040 The link to the dataset is in the paper. #NLProc #nlpmontclair #euphemisms @aclsigturk

🌟 How do large multimodal models perform on idiomatic expressions? 🌟 Current models often struggle with idioms due to their non-compositional nature. Advancing representations of these expressions is crucial to avoid errors in downstream tasks. 🧠📊 #MultimodalAI #Idioms #NLP

5/5 Thinking out loud: How can we enhance the cultural awareness and performance of LLMs across different cultures without access to the LLMs' training data? Our options are limited: fine-tuning with culturally appropriate datasets and prompt engineering.

Lab summer reading: @radamihalcea and her colleagues (aclanthology.org/2024.naacl-lon…) reveal gaps in AI's cultural understanding. Can LLMs truly grasp the nuances of different cultures? How should we tackle these biases in AI training? #AI #CulturalAwareness #LLM #BiasInAI