• Future Built on Knowledge
Machine Translation Unit
  • Home
  • People
  • Research
  • Projects
  • Publications
  • Resources
  • Join us
  • LOGIN

Select Page

Linee guida per la creazione del Gold Standard per la risoluzione della coreferenza di entità di tipo PERSONA in Adige-500K

MT Group at FBK Follow

#MachineTranslation Research Unit @FBK_research. #nlproc #deeplearning #ai

Avatar
Avatar MT Group at FBK @fbk_mt ·
12 Sep

Our pick of the week by @mgaido91: "Context-Driven Dynamic #Pruning for Large #Speech #Foundation Models" by Masao Someki, Shikhar Bharadwaj, Atharva Anand Joshi, Chyi-Jiunn Lin, Jinchuan Tian, Jee-weon Jung, @shinjiw_at_cmu, et al. (#INTERSPEECH2025).

Marco Gaido @mgaido91

as we are organizing the second edition of the IWSLT model compression task, happy to see new works on pruning large speech model based on external context (speaker, acoustic events, language)

https://arxiv.org/pdf/2505.18860

@fbk_mt

Reply on Twitter 1966518400294772807 Retweet on Twitter 1966518400294772807 Like on Twitter 1966518400294772807 1 Twitter 1966518400294772807
Avatar MT Group at FBK @fbk_mt ·
3 Sep

Our pick of the week by @FBKZhihangXie: "SimulMEGA: MoE Routers are Advanced Policy Makers for Simultaneous Speech Translation" by Chenyang Le, Bing Han, Jinshun Li, Songyong Chen, and Yanmin Qian (2025)

#Speech #Simultaneous #Translation #MOE #SpeechTech

Zhihang Xie @FBKZhihangXie

🚀 SimulMEGA: MoE Routers as advanced policy makers for Simultaneous Speech Translation 🎧🌍
Mixture-of-Experts routing → smarter decisions on when & how to translate, balancing latency vs quality in real-time speech. Paper link at https://arxiv.org/pdf/2509.01200v1

Reply on Twitter 1963187902277427545 Retweet on Twitter 1963187902277427545 2 Like on Twitter 1963187902277427545 2 Twitter 1963187902277427545
Avatar MT Group at FBK @fbk_mt ·
30 Aug

Our pick of the week by @beomseok_lee_: "#Speech Discrete Tokens or Continuous Features? A Comparative Analysis for Spoken Language Understanding in #SpeechLLMs" by Dingdong Wang, Junan Li, Mingyu Cui, Dongchao Yang, Xueyuan Chen, Helen Meng (#EMNLP2025)

#SLU #SpeechTech

Beomseok LEE @beomseok_lee_

🤔 Ever wondered how discrete tokens vs. continuous features behave in SpeechLLMs?
This new work dives into 6 SLU tasks and reveals some interesting takeaways!

https://arxiv.org/abs/2508.17863

Reply on Twitter 1961801977538494953 Retweet on Twitter 1961801977538494953 1 Like on Twitter 1961801977538494953 5 Twitter 1961801977538494953
Avatar MT Group at FBK @fbk_mt ·
21 Aug

Our pick of the week by @lina_conti: "I Have No Mouth, and I Must Rhyme: Uncovering Internal Phonetic Representations in LLaMA 3.2" Oliver McLaughlin, Arjun Khurana, @jack_merullo_ (2025)



#XAI #LLM #interpretability

Lina Conti @lina_conti

Pick of the week @fbk_mt : "I Have No Mouth, and I Must Rhyme" 👄 a dive into how LLaMA develops internal phonetic representations despite never hearing speech.
https://arxiv.org/pdf/2508.02527
#XAI #LLM #interpretability

Reply on Twitter 1958444375777702007 Retweet on Twitter 1958444375777702007 1 Like on Twitter 1958444375777702007 3 Twitter 1958444375777702007
Load More

Contact: dicenter-webmaster@fbk.eu
The use of content covered by intellectual property is carried out in compliance with articles 2, 21, 33 of the Constitution and art. 70 of the copyright law.
This publication was produced with the support of the Autonomous Province of Trento.