Dennis Fucci
@DennisFucci
Followers
183
Following
246
Media
3
Statuses
113
PhD student at @FBK_research - @fbk_mt group | Passionate about language and tech | Sea lover
Joined October 2021
@sarapapi @DennisFucci shared insightful findings on gender bias through the lens of interpretability: "Different speech translation models encode and translate speaker gender differently" ๐ https://t.co/4SpFnQdIzI (3/6)
1
1
6
Heading home after an exciting and intense @aclmeeting in Vienna! We had a great time presenting our work and connecting with the community. Thanks to everyone who came by! #acl2025 #nlproc (1/6)
5
3
28
Turn old photos into videos and see friends and family come to life. Try Grok Imagine, free for a limited time.
732
1K
5K
If you're in Rotterdam for Interspeech, join us on Monday for the special session on Interpretability in Speech Technology ๐ 11:00 AM ๐Dock 15 ๐๏ธย ๐๐๐ฒ๐ง๐จ๐ญ๐ by ๐๐๐๐. ๐พ๐๐๐๐ ๐ฟ๐๐ฃ๐๐ ๐๐ข ๐ผ๏ธ ๐๐จ๐ฌ๐ญ๐๐ซ ๐ฌ๐๐ฌ๐ฌ๐ข๐จ๐ง for 15 accepted papers #Interspeech2025
๐ค Just days before #Interspeech2025! We are excited to announce Prof. ๐๐ฎ๐ฟ๐ฒ๐ป ๐๐ถ๐๐ฒ๐๐ฐ๐ as our keynote speaker for the special session on Speech Interpretability! ๐ "What can interpretability do for us (and what can it not)?" ๐ More info https://t.co/ERvRCAcspQ
0
3
8
Findings from https://t.co/wFjhnZJwh3 show that speech SSL models converge on similar embedding spaces, but via different routes. While overall representations align, individual neurons learn distinct localized concepts. Interesting read! @fbk_mt
ieeexplore.ieee.org
Self-supervised models have revolutionized speech processing, achieving new levels of performance in a wide variety of tasks with limited resources. However, the inner workings of these models are...
0
1
3
๐ Last call for the Model Compression for Machine Translation task at #WMT2025 (co-located with #EMNLP2025)! Test data out on June 19 โก๏ธ 2 weeks for evaluation! Can you shrink an LLM and keep translation quality high? ๐ https://t.co/ZlDmwaZ90A
#NLP #ML #LLM #ModelCompression
0
9
8
๐ New tech report out! Meet FAMA, our open-science speech foundation model family for both ASR and ST in ๐ฌ๐ง English and ๐ฎ๐น Italian. The models are live and ready to try on @huggingface ๐ ๐ https://t.co/14PCvNsuCl
#ASR #ST #OpenScience #MultilingualAI
huggingface.co
``FAMA: The First Large-Scale Open-Science Speech Foundation Model for English and Italian,'' Sara Papi, Marco Gaido, Luisa Bentivogli, Alessio Brutti, Mauro Cettolo, Roberto Gretter, Marco Matassoni, Mohamed Nabih, Matteo Negri,
0
10
26
Pick of the week @fbk_mt: https://t.co/4x6EPOVZss by @soheunshim @DomenicoDeCris1. XAI work on cross-lingual alignment in speech-to-text models that disentangles phonetics and semantics. Plus: their XAI insights yield actionable improvements for low-resource language performance.
arxiv.org
Cross-lingual alignment in pretrained language models (LMs) has enabled efficient transfer in text-based LMs. Such an alignment has also been observed in speech foundation models. However, it...
0
2
5
๐ New shared task at #WMT2025 (co-located with @emnlpmeeting ): Model Compression for Machine Translation! Can you shrink an LLM and keep translation quality high?๐ง Submit by July 3 and push the limits of efficient NLP! ๐ https://t.co/ZlDmwaZ90A
#NLP #ML #LLM #ModelCompression
0
11
10
More great news! ๐ Our paper โEchoes of Phonetics: Unveiling Relevant Acoustic Cues for ASR via Feature Attributionโ was accepted at #Interspeech2025! Interested in interpretability for speech models? Preprint coming soon! โ๐ผ @mgaido91, @negri_teo, M.Cettolo, @luisabentivogli
0
6
15
๐ Excited to share our paper โDifferent Speech Translation Models Encode and Translate Speaker Gender Differentlyโ was accepted at #ACL2025 (main)! โ๐ผ Big thanks to amazing co-authors: @mgaido91, @negri_teo, @luisabentivogli, @andre_t_martins, @peppeatta! ๐ Preprint out soon!
0
7
29
๐ Excited to share that our @sarapapi has won the 2024 Best PhD Award from the Information and Engineering Doctoral School at @UniTrento_DISI for her thesis โDirect Speech Translation in Constrained Contexts: The Simultaneous and Subtitling Scenarios.โ #nlproc @FBK_research
1
6
20
๐ข Come and join our group! We offer a fully funded 3-year PhD position: ๐Automatic translation with large multimodal models https://t.co/jo587w3TQR ๐Full details for application: https://t.co/ZagIUM3mMQ ๐
Deadline May 12, 2025 #NLProc @FBK_research
0
10
10
Happy to announce that our paper 'An LLM-as-a-judge Approach for Scalable Gender-Neutral Translation Evaluation' was accepted at #GITT2025! ๐ Check it out: https://t.co/38efkWC9pZ ๐ฅ Co-authors (๐ซถ๐ป): @BeatriceSavoldi, @negri_teo, @luisabentivogli
arxiv.org
Gender-neutral translation (GNT) aims to avoid expressing the gender of human referents when the source text lacks explicit cues about the gender of those referents. Evaluating GNT automatically...
0
6
14
Yesterday I celebrated #WorldVoiceDay at @unito, sharing some thoughts and highlights from my research on AI and voice. A truly enjoyable event, with inspiring and diverse talks on the many dimensions of voice!
0
2
16
Our first guest speaker to join the virtual stage at #SlatorCon Remote is @FBK_researchโs @sarapapi, where she explores How โRealโ is Your Real-Time Simultaneous #Speech-to-Text #Translation System? ๐ก #SlatorConRemote #SCRemoteMar25 #AI #xl8 #t9n #NLP
1
6
17
Had a great time visiting @MilaNLProc last week and sharing some of my latest work on #XAI in #SpeechTranslation! Thanks for having me! ๐
๐For this Friday's @MilaNLProc lab seminar, it was a pleasure to have @DennisFucci presenting "Through the Lens of Explainability: Understanding Gender Bias in Speech Translation". #NLProc
0
1
12
As someone investigating how speech models use acoustic-related info, itโs great to see more work on evaluating whether prosodic features are leveraged in speech translation, like in https://t.co/ZYq1j6zGsw. We definitely need more research in this space! @fbk_mt
aclanthology.org
Ioannis Tsiamas, Matthias Sperber, Andrew Finch, Sarthak Garg. Proceedings of the Ninth Conference on Machine Translation. 2024.
0
0
7
๐ New @iwslt shared task on model compression: shrink a large multimodal model while preserving speech translation quality. Are you in? ๐ Smaller is greener! #IWSLT2025 #NLP #SpeechTech #SustainableAI
https://t.co/NK6U0krLYC
iwslt.org
Home of the IWSLT conference and SIGSLT.
0
7
14
We're on for Round 3! ๐ Working on #gender and #translationtechnology? We'll see you in Geneva at @MTSummit2025 for #GITT2025! ๐Submission deadline: 10 March ๐ธ๏ธInfo on our website: https://t.co/ouvLmrAbDg ๐Future updates at https://t.co/7smCpPN3co
0
8
10
I'm happy to share that our paper "Prepending or Cross-Attention for Speech-to-Text? An Empirical Comparison" has been accepted at @naacl @naaclmeeting 2025! #NAACL2025
@Lam19Tk @mgaido91 ๐ ๐ Preprint: https://t.co/7wi9WHU6MO โฐ Code will be released soon #NLProc #Speech
arxiv.org
Following the remarkable success of Large Language Models (LLMs) in NLP tasks, there is increasing interest in extending their capabilities to speech -- the most common form of communication. The...
1
7
41