🔍 Uppsala NLP Reading Group: Paper Spotlight! This week's pick: "In-Place Test-Time Training" by Feng. et al. 📖 LLMs can't adapt their weights at inference time. In-Place TTT fixes this by repurposing MLP blocks as "fast weights," enabling dynamic updates without retraining from scratch. Stay tuned!

🔍 Uppsala NLP Reading Group: Paper Spotlight! This week's pick: “Neural Morphological Tagging for Nguni Languages” by Marquard et al.. 📖 A study on neural morphological tagging for Nguni languages, showing strong results for task-specific neural models in low-resource settings. Let's discuss! 🚀

🔍 Uppsala NLP Reading Group: Paper Spotlight! This week's paper: “Large Language Models are Interpretable Learners” by Wang et al. The authors introduce LLM-Symbolic Programs to combine LLM reasoning with rule-based structures to produce interpretable models. Stay tuned for our next pick! 🚀

📢 Open positions at the department of Linguistics and Philology! 📢 Postdoktor i språkdokumentation och språkbeskrivning: lnkd.in/d2jmHaPZ Postdoktor i global filologi: lnkd.in/dKipcg7p Postdoktor i flerspråkig språkteknologi: lnkd.in/dAkSY6tW

📢 Open Positions at the Uppsala NLP Group! 📢 Postdoc opportunity — also open to recent or soon-to-be PhD graduates (within 1–2 months). uu.varbi.com/en/what:job/...

🔍 Uppsala NLP Reading Group: Paper Spotlight! This week's pick: “AI-LieDar: Examine the Trade-off Between Utility and Truthfulness in LLM Agents” by Su et al. 🤖 How should LLMs balance being helpful and being truthful in multi-turn interactions? Stay tuned for our next pick! 🚀

🔍 Uppsala NLP Reading Group: Paper Spotlight! This week's pick is “A novel unsupervised contrastive learning framework for ancient Yi script character dataset construction” by Bi, Sun & Chen. 🧠 New unsupervised method for better ancient Yi script datasets. 📖 Stay tuned for our next pick! 🚀

📢Life update📢 🥳I'm excited to share that I've started as a postdoc at Uppsala University NLP @uppsalanlp.bsky.social, working with Joakim Nivre on topics related to constructions and multilinguality! 🙏Many thanks to the Walter Benjamin Programme of the DFG for making this possible.

🔍 Uppsala NLP Reading Group: Paper Spotlight! This week's pick is "Do Prompt-Based Models Really Understand the Meaning of Their Prompts?" by Webson & Pavlick. 🧠 The authors ask whether modern LMs truly “understand” the instructions in prompts. 📖 Let’s discuss! Stay tuned for our next pick! 🚀

🔍 Uppsala NLP Reading Group: Paper Spotlight! This week's pick: "Modern Models, Medieval Texts: A POS Tagging Study of Old Occitan" by Schöffel et al. 🧠 Old Occitan exposes LLM struggles in POS tagging and points to tips for low-resource languages. 📖 Let’s discuss! Stay tuned for our next pick! 🚀

🔍 Uppsala NLP Reading Group: Paper Spotlight! This week’s pick: "Fantastically Ordered Prompts and Where to Find Them" by Lu et al. 🧠 Prompt order matters—a lot. This paper shows why and offers an unsupervised fix using entropy. 📖 Read it and let’s discuss! Stay tuned for our next pick! 🚀

🔍 Uppsala NLP Reading Group: Paper Spotlight! This week’s pick: "How Likely Do LLMs with CoT Mimic Human Reasoning?" by Bao et al. 🧠 CoT prompting can improve LLM reasoning—but does it mimic how humans think? 📖 Give it a read and share your thoughts! Stay tuned for our next pick! 🚀

@uppsalanlp.bsky.social and Språkbanken CLARIN are represented today at Svenska Språkets historia 18 #ssh18uu by @sstymne.bsky.social and Eva Pettersson

We just announced a new position as assistant professor / biträdande universitetslektor (BUL). For more information visit: uu.varbi.com/se/what:job/...

Biträdande universitetslektor i datorlingvistik

Uppsala universitet, Institutionen för lingvistik och filologi Institutionen för lingvistik och filologi bedriver forskning och utbildning inom språkvetenskap och textstudier med fokus på Asiens s

uu.varbi.com

🔍 Uppsala NLP Reading Group: Paper Spotlight! This week, we’re reading "Why do language models perform worse for morphologically complex languages?" by Arnett et al. 📖 Give it a read and share your thoughts! Stay tuned for our next pick! 🚀

🔍 Uppsala NLP Reading Group: Paper Spotlight! This week, we’re diving into "Leveraging Mandarin as a Pivot Language for Low-Resource Machine Translation between Cantonese and English" by Suen et al. 📖 Give it a read and share your thoughts! Stay tuned for our next pick! 🚀

🔍 Uppsala NLP Reading Group: Paper Spotlight! This week, we’re diving into "Uncertainty Modelling in Under-Represented Languages with Bayesian Deep Gaussian Processes" by Ubaid Azam et al. 📖 Give it a read and share your thoughts! What stood out to you? Let's discuss! Stay tuned for our next pick! 🚀

We have several open PhD positions (4-years, fully funded) at the Faculty of Languages at Uppsala University, both in Computational Linguistics and other language subjects. www.uu.se/en/about-uu/...

PhD positions at the Faculty of Languages - Uppsala University

PhD positions at the Faculty of Languages, Faculty of Languages, Uppsala University

uu.se

Ahmed will be presenting at virtual COLING today!

Christian Hardmeier@chrha.bsky.social · 2y ago

A new multimodal dataset of Arabic implicit discourse relations in speech and writing and a comparison of recognition methods based on audio and text - it's all in the paper by Ahmed Ruby, @sstymne.bsky.social and me, in Monday's online COLING session. Look here: aclanthology.org/2025.coling-...