Our #PickoftheWeek by @bsavoldi.bsky.social

Beatrice Savoldi@bsavoldi.bsky.social · last mo.

📖 #PickoftheWeek @fbk-mt.bsky.social "Accuracy: Community Perspectives on Machine Translation" A cool analysis of the conflicting interests of different communities around MT(AI developers, LSPs, and users) arxiv.org/pdf/2606.09655 #NLP #MachineTranslation #DiverseStakeholders

Last week, we had a great talk for our MT Seminar Series! @juliuscheng.bsky.social a PhD from the Cambridge NLP group presented a talk on "Effective uses of grammatical knowledge in extremely low-resource Machine Translation" #MachineTranslation #LowResourceMT #NLProc #FBK

We are at the Alliance for Language Technologies - #ALTEDIC Week 2026! @luisabentivogli.bsky.social and @matteo-negri.bsky.social are attending two project meetings (ALT-EDIC4EU and #LLMs4EU), presenting the Evaluation of Spoken Language Translation in the context of IWSLT. #LanguageTechnologies

We're excited to share that three papers from our lab have been accepted at #Interspeech2026 !! 🍾 #SpeechTranslation #SpeechAI #NLProc #FBK #Interspeech

Our #PickofTheWeek by @linaconti.bsky.social on Gender bias in Speech LLMs "Greater accessibility can amplify discrimination in generative AI" by Carolin Holtermann, Minh Duc Bui2, Kaitlyn Zhou, Valentin Hofmann, Katharina von der Wense, Anne Lauscher

Lina Conti@linaconti.bsky.social · 2mo ago

Pick of the week @fbk-mt.bsky.social: "Greater accessibility can amplify discrimination in generative AI" Gender bias in speech-based LLMs examined from multiple angles: a user survey, automatic bias measurement, and pitch manipulation experiments. arxiv.org/pdf/2603.22260

Late update, but we had two great talks last month! #MachineTranslation #FBK #NLProc #GenderBias #SpeechSynthesis From Harm Lameris, a PhD candidate at KTHuniversity on "Communicative Functions of Synthesized Speech: Modelling Prosody and Voice Quality Dynamics"

Our #PickofTheWeek by @uzumakidhairya.bsky.social: "Scaling Laws for Precision" by Tanishq Kumar, Zachary Ankner, Benjamin F. Spector, Blake Bordelon, Niklas Muennighoff, Mansheej Paul, Cengiz Pehlevan, Christopher Ré, Aditi Raghunathan (ICLR 2025)

@uzumakidhairya.bsky.social · 2mo ago

Pick of the week @fbk-mt.bsky.social Super interesting paper on the limitations of quantization, demonstrating how post-training quantization scales poorly in data. arxiv.org/abs/2411.04330

🏝️ Yesterday at #LREC2026, Palma de Mallorca! @linaconti.bsky.social presented "Voice, Bias, and Coreference: An Interpretability Study of Gender in Speech Translation" at the poster session. 📄Paper: arxiv.org/abs/2511.21517 💻Code: github.com/lina-conti/v... #SpeechTranslation #NLProc

Voice, Bias, and Coreference: An Interpretability Study of Gender in Speech Translation

Unlike text, speech conveys information about the speaker, such as gender, through acoustic cues like pitch. This gives rise to modality-specific bias concerns. For example, in speech translation (ST)...

arxiv.org

#PickOfTheWeek How does the granularity of speech-text pairs impact SpeechLLM performance, and what is the optimal way to interleave tokens? Furthermore, what are the best practices for generating synthetic data to boost training?🧐

⏰ 4 days left to apply! 🎓 2 PhD positions still open: 🎯 Human-Centred Evaluation Frameworks for Multilingual Technologies ✨ Multimedia Personalization with Multimodal Large Language Models 📅 Deadline: 15 May 2026 🔗 Full details: iecs.unitn.it/education/ad...

@linaconti.bsky.social & @luisabentivogli.bsky.social are heading to #LREC2026 in Palma! They'll present two papers: 📄 Voice, Bias, and Coreference: An Interpretability Study of Gender in Speech Translation 📄 Phonetic-based Ranking for Improved Pseudo-Labelling in Low-Resource ASR See you there! 🌴

Our #PickOfTheWeek by @lucacorbucci.bsky.social : "What Matters in Data for DPO? " by Yu Pan, Zhongze Cai, Guanting Chen, Huaiyang Zhong, Chonghuan Wang (NeurIPS 2025).

Luca Corbucci@lucacorbucci.bsky.social · 3mo ago

🤔 What Matters in Data for DPO? I asked myself this question a few days ago while trying to understand how to generate a dataset with preferences to run #DPO. This recent #NeurIPS paper answered some of my questions.

🎓 Come and join our group! 🎓 We offer 2 fully funded PhD positions: 🌍 Human-Centred Evaluation Frameworks for Multilingual Technologies (A6) 🤖 Multimedia Personalization with Multimodal Large Language Models (A7) ⏰ Deadline: 15 May 2026 🔗 Details: iecs.unitn.it/education/ad...

Our #PickOfTheWeek by @zhihangxie.bsky.social: "Detecting Hallucinations in SpeechLLMs at Inference Time Using Attention Maps" by Jonas Waldendorf, Bashar Awwad Shiekh Hasan, and Evgenii Tsymbalov (Findings of ACL 2026).

Zhihang Xie@zhihangxie.bsky.social · 3mo ago

🚀 New paper: Detecting Hallucinations in SpeechLLMs at Inference Time Using Attention Maps 📄 arxiv.org/abs/2604.19565 🧩 Lightweight inference-time detection for SpeechLLM hallucinations using audio-focused attention features. ✨ Attention classifiers outperform uncertainty baselines on ASR and S2TT.

🚀 New Shared Task: Model Compression for Machine Translation at hashtag#WMT2026 (co-located with hashtag#EMNLP2026)! 📅 Test data out on June 18th, submissions by July 2nd! Can you shrink an LLM and keep translation quality high? 🧠🔧 👉 www2.statmt.org/wmt26/model-... #NLP #ML #LLM #ModelCompression

At our last seminar, @maikezufle.bsky.social presented her work on Duplex Models, "Building Controllable Speech Systems"

🎉We’re happy to welcome our new research engineer, Yiu Chung Leung, joining us to work on multimodal LLMs with a focus on speech. Excited for what’s ahead! 🚀

With @malvinanissim.bsky.social and @vivipatti.bsky.social, we've been teaching ethics in NLP as a hands-on course across Groningen, Pavia & Turin. We wrote up the experience and received the ✨Best Paper Award✨ at #EACL2026's TeachNLP Workshop. Huge thanks to the organizers and all our students!

📢Thrilled to welcome our new postdoc, @wafaaissa.bsky.social! Looking forward to working together on trustworthy and responsible language technologies 🚀

For our seminars at FBK, last week we had a great talk by @gsarti.com on Interpreting Context Usage in Generative Language Models

🎉 We’re very happy to welcome our new postdoc @lucacorbucci.bsky.social, who will be working on multimodal LLMs. Looking forward to the exciting research ahead! 🚀

Will #GITT2026 be worth it? If our brilliant keynotes are anything to go by...😉 Introducing @odettes.bsky.social Full Professor of Inclusive Speech Communication at 🇳🇱 Delft University of Technology. Human & automatic speech, combining technical x ethical x societal, just how we like it! Excited yet?

At our last MT seminar, we had @jacquelinerowe.bsky.social from @edinburghuni.bsky.social presenting her work "Beyond Benchmarks? Measuring Gender Bias in Multilingual LLMs"

🎓 Today we attended the FBK PhD Day, and our amazing students presented their research lines: @apierg.bsky.social: gender-inclusive MT @linaconti.bsky.social: explainable speech translation @zhihangxie.bsky.social: long-form SpeechLLMs @uzumakidhairya.bsky.social: resource-efficient translation

Our #PickOfTheWeek by @bsavoldi.bsky.social: "Attention to Non-Adopters" by @kaitlynzhou.bsky.social, @gligoric.bsky.social, @myra.bsky.social, @mlam.bsky.social, @vyoma-raman.bsky.social, Boluwatife Aminu, Caeley Woo, Michael Brockman, @hannah-cha.bsky.social, @jurafsky.bsky.social (2025).

Beatrice Savoldi@bsavoldi.bsky.social · 6mo ago

🗞️ Pick of the week @fbk-mt.bsky.social: Most LLM dev centers current adopters, so what are we missing? Worth a read! 👇https://arxiv.org/pdf/2510.15951 #NLP #LLMs #HCI

Great seminar last week by Alessandra Teresa Cignarella @cerveza-inglesa.bsky.social "6,000 Papers, One New Dataset, and Several Confused LLMs: A Story About Stereotype Detection in NLP" #NLP #StereotypeDetection

🗞️ Pick of the week @fbk-mt.bsky.social: Most LLM dev centers current adopters, so what are we missing? Worth a read! 👇https://arxiv.org/pdf/2510.15951 #NLP #LLMs #HCI

Last week at the @fbk-mt.bsky.social seminars, we hosted Elizabeth Salesky from Google DeepMind, presenting her work on "Translation and Language Modeling with Pixels" #NLProc #tokenization #MT

Our pick of the week by @linaconti.bsky.social: "Voice gender diversity: expression, perception and acoustics" by @victor-rosi.bsky.social and Carolyn McGettigan, Royal Society Open Science (2025)

Lina Conti@linaconti.bsky.social · 7mo ago

💬 Pick of the week @fbk-mt.bsky.social: "Voice gender diversity: expression, perception and acoustics." A great review of gender-diverse voices. Covers acoustic properties, perception and the gap between self-expression of gender and how voices are heard. royalsocietypublishing.org/rsos/article...

Our pick of the week by @apierg.bsky.social: "Glitter: A Multi-Sentence, Multi-Reference Benchmark for Gender-Fair German Machine Translation" by Praval, @jhackenbuchner.bsky.social, @gattanasio.cc, and @a-lauscher.bsky.social (EMNLP 2025 Findings)

Andrea Piergentili@apierg.bsky.social · 7mo ago

Impressive work by the Glitter team: a new human-made benchmark for German gender-inclusive MT with long passages and multiple inclusive approaches + experiments showing that MT systems and LLMs still fall short in generating inclusive outputs. aclanthology.org/2025.finding... ✨