This work was done in collaboration between
@copenlu.bsky.social and UILab. Cannot thank the co-authors, Seogyeong Jeong, @siddheshp.bsky.social, Jisu Shin, Jiho Jin, Junho Myung, and supervisors @iaugenstein.bsky.social, Alice Oh enough!
16.08.2025 12:31 โ ๐ 2 ๐ 0 ๐ฌ 0 ๐ 0
๐ What we found
1๏ธโฃ LLMs encode Western-dominance bias and cultural flattening in their internals.
2๏ธโฃ Internalized cultural biases do not necessarily align with extrinsic biases.
3๏ธโฃ Low-resource cultures are less affected by these biases, likely due to limited training data.
16.08.2025 12:31 โ ๐ 2 ๐ 0 ๐ฌ 1 ๐ 0
๐ How do Large Language Models internally process cultural knowledge? ๐
Happy to share our new preprint "Entangled in Representations: Mechanistic Investigation of Cultural Biases in Large Language Models"
๐ Paper: arxiv.org/abs/2508.08879
16.08.2025 12:31 โ ๐ 10 ๐ 3 ๐ฌ 1 ๐ 0
Post-doctoral Researcher at BIFOLD / TU Berlin interested in interpretability and analysis of language models. Guest researcher at DFKI Berlin. https://nfelnlp.github.io/
PhD student @mainlp.bsky.social (@cislmu.bsky.social, LMU Munich). Interested in language variation & change, currently working on NLP for dialects and low-resource languages.
verenablaschke.github.io
PhD Student at the ILLC / UvA doing work at the intersection of (mechanistic) interpretability and cognitive science. Current Anthropic Fellow.
hannamw.github.io
Strengthening Europe's Leadership in AI through Research Excellence | ellis.eu
Professor at the University of Copenhagen. Explainable AI, Natural Language Processing, ML. Head of copenlu.bsky.social lab.
#NLProc #NLP #XAI
http://isabelleaugenstein.github.io/
Assistant Professor at Bar-Ilan University
https://yanaiela.github.io/
Asst prof @ University of Utah ยท NLP ยท she/her ๐ญ๐ท
language model research @ai2.bsky.social, Co-lead of Data for OLMo w/ @soldaini.net, statistics @uw, open science, tabletop, seattle, he/him,๐ง kyleclo.com
The NLP group at the University of Washington.
NLP PhD student at Chalmers University of Technology, Sweden. Working on retrieval augmented language models and interpretability.
Research collaboration among 5 universities in Denmark: Aalborg University, IT University of Copenhagen, University of Aarhus, Technical University of Denmark, and the University of Copenhagen.
https://www.aicentre.dk/
PhD student working on NLP, LLM, AI, GNN, Drug Discovery and something more ...
Postdoc at the University of Copenhagen interested in human-AI interaction ๐ฉโ๐ป, explainable AI ๐ & NLP ๐.
Currently researching human-centred explainable fact-checking.
๐ https://gretawarren.github.io/
๐ ๐ฉ๐ฐ/๐ฎ๐ช
EU Policy Lead & Applied Researcher @ Hugging Face ๐ค
Computer Scientist, PhD
Wikipedia & languages are my โก
Postdoc at the University of Tรผbingen, supervised by @andreasgeiger.bsky.social | Research: Scientific Document Analysis, Cultural AI, & LLM Agents.
Assistant Professor at Department of Computer Science, University of Copenhagen. Cross-cultural Natural Language Processing, food and positive impact are my passion. Dad, kendoka and language learner. danielhers.github.io
Grad Student; Into Multilingual NLP
Researcher @Nvidia | PhD from @CopeNLU | Formerly doing magic at @Amazon Alexa AI and @ARM. ML MSc graduate from @UCL. Research is the name of the game. แแแข
http://osoblanco.github.io
PhD Fellow | University of Copenhagen (CopeNLU) | Pioneer Centre for AI
Sarah Masud, Postdoc at CopeNLU.
https://sara-02.github.io
