Jannis Vamvas

Jannis Vamvas

@vamvas.bsky.social

Natural Language Processing (#NLProc #NLP) Researcher @ University of Zurich @cl-uzh.bsky.social vamvas.ch

156 followers100 following11 posts

Jannis Vamvas reposted

Computational Linguistics @ UZH
@cl-uzh.bsky.social

Only a few more days left to register for SwissText 2026 with early-bird rates!

Computational Linguistics @ UZH@cl-uzh.bsky.social

🗓️ SwissText 2026 keynote speakers announced & registration open! We are delighted to welcome Prof. Dr. Alexandra Birch and Dr. Valentina Pyatkin as our keynote speakers. 📋 Register here: ema.uzh.ch/RHK4W Early-bird rates available throughout April, with additional student discounts. #NLProc 1/3

ema.uzh.chSwissText 202610. Juni 2026 | UZH Campus Oerlikon

Jannis Vamvas reposted

Rico Sennrich
@ricosennrich.bsky.social

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...

jobs.uzh.chUZH: PhD in Language AI / Natural Language ProcessingYou 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 ...

Jannis Vamvas reposted

Hanxu Hu
@hanxuhu.bsky.social

💥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)

View full thread
Jannis Vamvas
@vamvas.bsky.social

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

Jannis Vamvas
@vamvas.bsky.social

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

Jannis Vamvas
@vamvas.bsky.social

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

Jannis Vamvas
@vamvas.bsky.social

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...

jobs.uzh.chUZH: PhD Student in Natural Language ProcessingYou 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...
Jannis Vamvas
@vamvas.bsky.social

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

github.comGitHub - ZurichNLP/mbr: Minimum Bayes Risk Decoding for Hugging Face TransformersMinimum Bayes Risk Decoding for Hugging Face Transformers - GitHub - ZurichNLP/mbr: Minimum Bayes Risk Decoding for Hugging Face Transformers

You are up to date.