Danni Liu
dnliu.bsky.social
Danni Liu
@dnliu.bsky.social
PhD student at Karlsruhe Institute of Technology, Germany. Working on multilinguality and machine translation.
🔗 https://danni-liu-cv.github.io
Reposted by Danni Liu
🚀 We're hiring a Researcher in Responsible & Trustworthy NLP! Join our research group @fbk-mt.bsky.social at Fondazione Bruno Kessler to work on fairness and trustworthiness in multilingual technologies.

📅 Deadline: Dec 10, 2025
🔗 Apply: jobs.fbk.eu/Annunci/Offe...
Jobs | Science and Technology Hub - Trento | A Researcher in Responsible and Trustworthy NLP
jobs.fbk.eu
November 25, 2025 at 9:07 AM
Reposted by Danni Liu
Excited to head over to #Suzhou to present 5 papers at #EMNLP2025 and affiliated venues!
Topics include quality estimation and evaluation 👩‍🔬, speech🗣️, and multilinguality🌐- see you soon! 🤩
@maikezufle.bsky.social @jan-niehues.bsky.social
October 31, 2025 at 1:34 PM
Reposted by Danni Liu
Ich freue mich schon auf den Vortrag und Austausch!
Wir bei AI4LT arbeiten täglich mit 🤖 KI und 🎤 Sprache. Aber wie funktioniert Künstliche Intelligenz eigentlich – und wie kann sie Texte und gesprochene Sprache erzeugen?
@maikezufle.bsky.social aus unserer Gruppe erklärt es in einem Vortrag in der Stadtbücherei #Wörth 📚.
English version below👇
October 23, 2025 at 9:40 AM
Reposted by Danni Liu
What if we did a single run and declared victory
October 23, 2025 at 2:28 AM
Reposted by Danni Liu
Hi all, there is a postdoc position open in the group I'm currently based in! ✨ Let me know if you are interested or have questions 🙂 Please share if you know someone who might be interested www.uu.nl/en/organisat...
Postdoctoral Researcher in Memory access in language
Help uncover how memory shapes language use. As a postdoctoral researcher at the Institute for Language Sciences, you will join the ERC-funded MEMLANG project.
www.uu.nl
October 8, 2025 at 7:55 AM
Reposted by Danni Liu
Writing: just add coffee
**Hey, my new book comes out on Tuesday! 👉 amazon.com/gp/aw/d/1419764063
September 14, 2025 at 9:15 PM
interesting read :) 📌
The main ingredient that led to GRPO's performance leap is the calibration of the reward/value via multiple rollouts per prompt.

Let me elaborate on what I mean by that and a cheaper way of doing it offline.
August 12, 2025 at 8:31 AM
Reposted by Danni Liu
Sara Papi, Maike Z\"ufle, Marco Gaido, Beatrice Savoldi, Danni Liu, Ioannis Douros, Luisa Bentivogli, Jan Niehues
MCIF: Multimodal Crosslingual Instruction-Following Benchmark from Scientific Talks
https://arxiv.org/abs/2507.19634
July 29, 2025 at 9:12 AM
Reposted by Danni Liu
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
https://arxiv.org/abs/2508.04796
August 8, 2025 at 7:12 AM
Reposted by Danni Liu
Let the games begin! It's time to watch eight LLMs compete in the first round of head-to-head match-ups. #KaggleGameArena
August 5, 2025 at 6:03 PM
Reposted by Danni Liu
Newton's Laws of Graduation, Part 1
August 2, 2025 at 11:36 PM