🔵 Michelle Wastl, Jannis Vamvas, Rico Sennrich. SwissGov-RSD: A Human-annotated, Cross-lingual Benchmark for Token-level Recognition of Semantic Differences Between Related Documents. 🔵 Zifan Jiang, Youngjoon Jang, Liliane Momeni, Gül Varol, Sarah Ebling, Andrew Zisserman. Segment, Embed, and Align: A Universal Recipe for Aligning Subtitles to Signing. 🔵 Kevin Du, Clara Kümpel, Michelle Wastl, Alex Warstadt. It’s Not What You Say, It’s How You Say It: Evaluating LLM Responses to Expressions of Belief. 🔵 Negar Foroutan, Clara Meister, Debjit Paul, Joel Niklaus, Sina Ahmadi, Antoine Bosselut, Rico Sennrich. Parity-Aware Byte-Pair Encoding: Improving Cross-lingual Fairness in Tokenization. 🔵 Apertus: Democratizing Open and Compliant LLMs for Global Language Environments. https://lnkd.in/eFTh-5m7 🔵 (Findings:) Elias Schuhmacher, Andrianos Michail, Juri Opitz, Rico Sennrich, Simon Clematide. Information Representation Fairness in Long-Document Embeddings: The Peculiar Interaction of Positional and Language Bias.

Looking forward to the ACL 2026 conference in San Diego, California!
Several accepted papers have involvement from our department:

1/7

UZH: PhD in Language AI / Natural Language Processing You will be joining the Department of Computational Linguistics, which has 6 Research Groups and around 70 postdoctoral and student researchers in the areas of Text Technologies, Phonetics and Speech ...

My lab is recruiting one PhD student and one post-doctoral researcher for a start as soon as this Summer! Apply by April 1 / March 31 to be among first candidates considered.

jobs.uzh.ch/job-vacancie...

jobs.uzh.ch/job-vacancie...

📣 SwissText 2026 – Call for Papers is open!

🎯 Special theme: #ReproducibleNLP
📅 Submission deadline: 17 March 2026
📍 Zurich, Switzerland · 10 June 2026

www.swisstext.org/call-for-pap...

(1/4)

💥Introducing new paper: arxiv.org/pdf/2510.17715, QueST — train specialized generators to create challenging coding problems.
From Qwen3-8B-Base
✅ 100K synthetic problems: better than Qwen3-8B
✅ Combining with human written problems: matches DeepSeek-R1-671B
🧵(1/5)

Sina Ahmadi receiving award.

Congratulations to @sinaahmadi.bsky.social and co-authors for receiving an ACL 2025 Outstanding Paper Award for PARME: Parallel Corpora for Low-Resourced Middle Eastern Languages!

aclanthology.org/2025.acl-lon...

Sina Ahmadi, Rico Sennrich, Michelle Wastl, Jannis Vamvas, Andrianos Michail from the Department of Computational Linguistics posing in front of the at ACL 2025 Vienna step-and-repeat banner

We're proud to see our researchers sharing their latest work: @sinaahmadi.bsky.social, @ricosennrich.bsky.social, @michellewastl.bsky.social, @vamvas.bsky.social and Andrianos Michail (pictured) and many others.
If you’re at #ACL, stop by to learn more!

Arrived in Vienna ready for #ACL2025NLP!

Excited to present papers with @vamvas.bsky.social @ricosennrich.bsky.social on Unsupervised Translation Direction Detection and Multilingual Hallucination Detection!

Come say hi! 👋
#NLProc #NLP #NMT #LLMs

Make a guess: How does GPT-4o tokenize the Lithuanian word ‘nebeprisikiškiakopūsteliautum’? A. ⟨nebepr, is, ik, iškiak, opūste, liautum⟩ B. ⟨neb, pre, ski, kayak, opus, tell, autumn⟩ C. ⟨ne, be, pris, iki, ški, ak, op, ū, stel, ia, utum⟩

Can you solve my exam question?
I've come to believe that multiple-choice exams are underrated. More in my blog post, “The Joy of Multiple-Choice.” vamvas.ch/the-joy-of-m...

I had the pleasure of presenting our paper “20min-XD: A Comparable Corpus of Swiss News Articles (arxiv.org/abs/2504.21677) ”, a joint work with @vamvas.bsky.social , Selena Calleri and @ricosennrich.bsky.social , at SwissText 2025 last week.

We are honored to receive Best Paper Award for it! ✨

Michelle Wastl at the SwissNLP Expo

Our Michelle Wastl is demonstrating her invention, the Translation Direction Detector, at the SwissNLP Expo!
Michelle's paper: arxiv.org/abs/2401.06769
Demo: huggingface.co/spaces/Zuric...
If you're at the expo, make sure to stop by the Department of Computational Linguistics UZH!

Bar chart illustrating counts of different API responses for the same request.

I also found what others complained about previously: The “Predicted Outputs” feature is unstable even with temp=0.

It’s unclear to me why this happens – do they batch the verification steps of users? An artifact of temperature sampling?

So, any hints would be appreciated! 4/4

3D plot comparing API behavior to simulation results.

Systematic probing of the OpenAI API reveals that they are using a lookahead parameter of K=16, and perform prefix matching with a prefix threshold of 32 tokens.

A simulation experiment confirms that the reverse-engineered algorithm matches the behavior of the API. 3/4

Schematic illustration of draft verification

Draft verification is a well-known idea in NLP, from Speculative Decoding to algorithms leveraging a static draft, such as Aggressive Decoding and Prompt Lookup Decoding.

Clearly, “Predicted Outputs” does something similar. But how does it work, specifically? 2/4

OpenAI Platform Explore developer resources, tutorials, API docs, and dynamic examples to get the most out of OpenAI's platform.

OpenAI introduced the “Predicted Outputs” feature last November, but their documentation raises more questions than it answers. So let’s reverse-engineer the feature! 🧵1/4

platform.openai.com/docs/guides/...

OpenAI's Speculative Decoding, Reverse-Engineered Why LLMs are faster if we give them a draft to complete.

Blogged: Reverse-engineering OpenAI's Speculative Decoding.

I was curious how GPT-4o can make use of predicted outputs to speed up text generation.
vamvas.ch/openai-predi...

Congratulations to Dr. @shaitarova.bsky.social , who just successfully defended her thesis on "Generated Text as the Locus of Language Contact: Exploring the Impact of Generative AI on English and German". Special thanks to the external examiner Natalia Resende!

I'm thrilled to share my first PhD project, a joint work with
@vamvas.bsky.social and @ricosennrich.bsky.social
Paper link:
arxiv.org/pdf/2503.10494
Long context LLMs have paved the way for document translation, but is simply inputting the whole content the optimal way?
Here's the thread 🧵 [1/n]

Tannon Kew presenting during his viva.

Congratulations to Dr. @tannonk.bsky.social, who just successfully defended his thesis on "Leveraging Data, Decoding, and Context for Controlling Text Generation from Pretrained Language Models". Special thanks to the external examiner @feralvam.bsky.social!

UZH: PhD Student in Natural Language Processing You will be joining the Department of Computational Linguistics, which has 5 faculties and around 40 postdoctoral and student researchers in the areas of text technologies, phonetics and speech. You w...

Rico Sennrich (University of Zurich) is hiring a PhD student to work on representations of knowledge in multilingual language models, starting September 2024.

Rico is my former advisor and I can greatly recommend working with him. Apply by January 4: jobs.uzh.ch/offene-stell...

GitHub - ZurichNLP/mbr: Minimum Bayes Risk Decoding for Hugging Face Transformers Minimum Bayes Risk Decoding for Hugging Face Transformers - GitHub - ZurichNLP/mbr: Minimum Bayes Risk Decoding for Hugging Face Transformers

Minimum Bayes Risk decoding has recently gained momentum in #NLProc.

However, a generic implementation for @huggingface.bsky.social Transformers has been missing. Check out our new 𝗺𝗯𝗿🔥 repo, which is designed to work with any model and metric on the Hub: github.com/ZurichNLP/mbr

Paper page - Towards Unsupervised Recognition of Semantic Differences in Related Documents Join the discussion on this paper page

Can NLP be used for highlighting differences between two documents?
Our EMNLP paper investigates the task of Recognizing Semantic Differences (RSD) with simple, unsupervised approaches.

• Demo: huggingface.co/spaces/Zuric...
• Paper: huggingface.co/papers/2305....

#NLProc