On May 12, @a-lauscher.bsky.social gives a keynote at the Social Context workshop: socon-nlpsi.github.io

A full week for the lab. If you are at #LREC2026, come find us.

We are also co-organizing the Identity-Aware AI workshop on May 16 (morning session). If your work sits at the intersection of identity, fairness, and NLP, come join us. #LREC2026 identity-aware-ai.github.io

Our paper with @queerinai.com is accepted at the Identity-Aware AI workshop: "Queering the Audits: Community-Based Auditing of AI Harms to Queer Communities." featuring @pranav-nlp.bsky.social and
@a-lauscher.bsky.social as authors.

"Detecting Hallucinations in Authentic LLM-Human Interactions", studying hallucinations in real LLM-human conversations rather than synthetic benchmarks. Work by
@yujie-ren.bsky.social, Niklas Gruhlke, @a-lauscher.bsky.social.

"Issue Detection and Category Classification in Domain-Specific Technical Logbooks", a collaboration with DESY on NLP for scientific infrastructure. Presented by the authors: Afshin Karimi, Ingmar Hartl, and Henrik Tunnermann,
@a-lauscher.bsky.social.

The Trustworthy AI Lab is heading to #LREC2026 in Palma de Mallorca this May with two papers, a workshop paper, a workshop we co-organize, and a keynote. Here is what to expect.

Some happy news 🎉 I’m happy to share that I’ve been promoted to Full Professor of Trustworthy AI in the Department of Informatics at the University of Hamburg. Very grateful for the amazing team and collaborators who made this possible. @tai-lab.bsky.social Looking forward to the next years!

🚨 New paper alert: A new generation of LLMs can now process speech natively. This could expand access for millions excluded by text interfaces, but our research shows a cost: demographic cues in speaker voice can trigger stereotypical model responses. 🎙️⚖️

Paper: arxiv.org/abs/2603.22260

Join us!! 😊

🎓 𝗣𝗵𝗗 𝗱𝗲𝗳𝗲𝗻𝘀𝗲 𝗰𝗼𝗻𝗴𝗿𝗮𝘁𝘂𝗹𝗮𝘁𝗶𝗼𝗻𝘀 𝘁𝗼 𝗔𝗻𝗱𝗿𝗲𝗮𝘀 𝗪𝗮𝗹𝗱𝗶𝘀!

Congratulations to @tresiwald.bsky.social on the successful defense of his dissertation “𝘈𝘯 𝘐𝘯𝘵𝘦𝘨𝘳𝘢𝘭 𝘝𝘪𝘦𝘸 𝘰𝘯 𝘵𝘩𝘦 𝘙𝘦𝘭𝘪𝘢𝘣𝘪𝘭𝘪𝘵𝘺 𝘰𝘧 𝘓𝘢𝘯𝘨𝘶𝘢𝘨𝘦 𝘔𝘰𝘥𝘦𝘭𝘴 𝘧𝘰𝘳 𝘊𝘰𝘮𝘱𝘶𝘵𝘢𝘵𝘪𝘰𝘯𝘢𝘭 𝘈𝘳𝘨𝘶𝘮𝘦𝘯𝘵𝘢𝘵𝘪𝘰𝘯”, held at Technische Universität Darmstadt on February 17th, 2026.

Jakob Prange from UKE presented on "The Language of Addiction": a computational linguistics perspective on child and adolescent psychiatry. Exciting to see interdisciplinary research bridging computational linguistics and mental health.

Great talk at our Data Science Lab this week!

Jakob Prange from UKE presented on "The Language of Addiction": a computational linguistics perspective on child and adolescent psychiatry. Exciting to see interdisciplinary research bridging computational linguistics and mental health.

Deadline extension alert! Identity-aware AI #LREC2026 extended to February 25th! All info at: identity-aware-ai.github.io

Deadline approaching! Identity-aware AI workshop #LREC2026 identity-aware-ai.github.io, there is time until Feb 20th to submit a short or full paper, with many non-archival options too!

For our weekly reading group, Afshin Karimi presented "What Social Media Use Do People Regret? An Analysis of 34K Smartphone Screenshots with Multimodal LLM". Super interesting discussion!

Paper: dl.acm.org/doi/10.1145/...

#NLProc

Holidays over, reading group resumes 📖
Today Henning Hoffmann presented the paper "Music for All: Representational Bias and Cross-Cultural Adaptability of Music Generation Models"

Paper: arxiv.org/pdf/2502.07328

#NLProc

This week at reading group 📚
@pranav-nlp.bsky.social presented "You Cannot Sound Like GPT": Signs of language discrimination and resistance in computer science publishing.

Paper: arxiv.org/abs/2505.08127

#NLProc

Ausschreibung

Benefits: 30 days vacation, flexible hours, work-life balance, health programs (EGYM), pension, professional development.

Hamburg is a vibrant, international city with a great quality of life!

Full details & application: www.uni-hamburg.de/stellenangeb...

We're looking for researchers passionate about:

Ethical & safe generative AI
Multilingual/multicultural AI
Multimodality & agentic systems
Model interpretability

Strong pub record in *CL/AI/ML/FAccT conferences etc. valued. We're a dynamic, well-connected team committed to diversity!

Exciting postdoc opportunity in Trustworthy AI & NLP with @a-lauscher.bsky.social!

Work on cutting-edge generative AI while addressing ethics, fairness, multilingualism & sustainability. 3-year position, teaching included.

Deadline: March 2
Start: June 1, 2026

Submission deadline: February 20, 2026
Notification: Late March
Workshop: May 16 (hybrid format)

We look forward for your submissions!

You can submit archival papers (4-8 pages) that get published in LREC proceedings, non-archival versions if your paper is published elsewhere, or creative formats like art, poetry, TikToks, blogs, and teaching materials.

We're looking for works on bias detection and fairness interventions, perspectivism, policymaking, auditing, and related topics that center identity in NLP.

AI is developed in a way that amplifies the majority. Current approaches harm marginalized communities and ignore identity in AI systems. This workshop challenges the harmful assumption that NLP should treat all humans as the same.

We are organizing the Identity-Aware NLP Workshop at LREC 2026 (Palma de Mallorca, May 16). #LREC2026

More info: identity-aware-ai.github.io

Decision-Making with Deliberation: Meta-reviewing as a Document-grounded Dialogue by @sukannya.bsky.social, @nilsdy.bsky.social, @a-lauscher.bsky.social, and @igurevych.bsky.social.

SoS: Analysis of Surface over Semantics in Multilingual Text-To-Image Generation by @carolin-holtermann.bsky.social, @floschne-nlp.bsky.social, @a-lauscher.bsky.social.

TempViz: On the Evaluation of Temporal Knowledge in Text-to-Image Models by @carolin-holtermann.bsky.social, Nina Krebs, and @a-lauscher.bsky.social.

Glad to announce that three papers from our group got accepted to #EACL2026! The details of the papers are in the thread:

Check out the paper here: arxiv.org/abs/2509.13835

✨Congratulations to all the amazing authors of this work ✨: @minhducbui.bsky.social @carolin-holtermann.bsky.social @valentinhofmann.bsky.social @a-lauscher.bsky.social

“You speak Bavarian? Then you must be uneducated + closed-minded.”

🤯 Not your opinion? Good. But it might be your LLM’s!!

🧵 Check out our #EMNLP2025 paper, where we uncover concerning dialect bias in recent LLMs - including GPT-5.

#AI #Bias #Dialect #Fairness #LLM #NLProc #Safety