Reposted by Jan Niehuesmaikezufle.bsky.social @maikezufle.bsky.social · 17/07/2026Join our Last Translation Benchmark Project! 😍 031
Reposted by Jan NiehuesMT Group at FBK @fbk-mt.bsky.social · 17/06/2026We 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 032
Reposted by Jan NiehuesMT Group at FBK @fbk-mt.bsky.social · 11/06/20261️⃣ "Do What I Say: A Spoken Prompt Dataset for Instruction-Following" 👥 @maikezufle.bsky.social, @sarapapi.bsky.social, Fabian Retkowski, Szymon Mazurek, Marek Kasztelnik, Alexander Waibel, @luisabentivogli.bsky.social, @jan-niehues.bsky.social 🇪🇺 Meetween EU project 📄 arxiv.org/abs/2603.09881arxiv.orgDo What I Say: A Spoken Prompt Dataset for Instruction-FollowingSpeech Large Language Models (SLLMs) have rapidly expanded, supporting a wide range of tasks. These models are typically evaluated using text prompts, which may not reflect real-world scenarios where ... 156
Reposted by Jan NiehuesKIT-Fakultät für Informatik @kitinformatik.bsky.social · 28/05/2026🎉 Big congrats to Maike Züfle from our AI4LT research group! She’s been selected for the Apple Scholars in AIML PhD Fellowship 🍎🤖 In this clip, Maike shares what she’s working on at KIT and what drives her research. 👩💻✨ @ai4lt.bsky.social @maikezufle.bsky.social @jan-niehues.bsky.social 032
Reposted by Jan NiehuesVilém Zouhar @zouhar.bsky.social · 30/04/2026Deadline extension! - The task is simple: get audio + its translation and estimate how good it is. - Mark your name as the winner of the first Speech Translation Metrics Shared Task at IWSLT 2026 🏆 Predictions submission: May 7, 2026 Description paper: May 10, 2026 122
Reposted by Jan Niehuesmgaido91.bsky.social @mgaido91.bsky.social · 15/04/2026🚀 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 #ModelCompressionlnkd.inLinkedInThis link will take you to a page that’s not on LinkedIn 065
Reposted by Jan NiehuesMT Group at FBK @fbk-mt.bsky.social · 09/04/2026At our last seminar, @maikezufle.bsky.social presented her work on Duplex Models, "Building Controllable Speech Systems" 087
Jan Niehues @jan-niehues.bsky.social · 25/03/2026Sai is currently presenting BOOM. If you are at EACL, drop by to see multimodal translation. 030
Reposted by Jan NiehuesAI4LT @ai4lt.bsky.social · 24/03/2026We are at @eaclmeeting.bsky.social in Rabat! Check out our work and collaborations: 📊 "Early-Exit and Instant Confidence Translation Quality Estimation" — Poster Session 2 (Wed 14:30) 🚀 "BOOM: Beyond Only One Modality — KIT's Multimodal Multilingual Lecture Companion" — Demo Session 1 (Wed 11:30) 041
Reposted by Jan NiehuesMatteo Negri @matteo-negri.bsky.social · 02/03/2026🚀 Call for Participation: @iwslt Offline Speech Translation 2026 Break language barriers with new languages & real-world scenarios + a brand new source-language agnostic speech translation track 🌍 📅 Evaluation: Apr 1–15 👉 iwslt.org/2026/offline #IWSLT2026 #SpeechAIiwslt.orgOffline ST trackHome of the IWSLT conference and SIGSLT. 051
Reposted by Jan NiehuesMatteo Negri @matteo-negri.bsky.social · 02/03/2026🚀 Call for Participation: @iwslt Model Compression 2026 Make large multilingual foundation models small ⚡ without losing power in EN→DE/ZH speech-to-text translation. 📅 Evaluation: Apr 1–15 iwslt.org/2026/compres... #IWSLT2026 #SpeechAI #Qwen2 #EfficientAIiwslt.orgCompression trackHome of the IWSLT conference and SIGSLT. 021
Reposted by Jan NiehuesMatteo Negri @matteo-negri.bsky.social · 02/03/2026🚀 Call for Participation: @iwslt Subtitling 2026 Turn speech into ready-to-watch subtitles 🎬 across TV, News & YouTube! 📅 Evaluation: Apr 1–15 iwslt.org/2026/subtitl... #IWSLT2026 #SpeechAI #MultimodalAIiwslt.orgSubtitling trackHome of the IWSLT conference and SIGSLT. 021
Reposted by Jan NiehuesAI4LT @ai4lt.bsky.social · 17/02/2026📢 Fully-funded #PhD position in #NLProc, focusing on #SpeechLLMs & multimodal language processing. Reposts appreciated! 🙏 📅 Apply by March 15! 🔗 ai4lt.iar.kit.edu/english/team... 045
Reposted by Jan NiehuesVilém Zouhar @zouhar.bsky.social · 09/02/2026Quality estimation (automated metrics) are amazing. Truly. We would like to use them everywhere. That gets compute-expensive very quickly. We also don't know when they don't know. In "Early-Exit and Instant Confidence Translation Quality Estimation" (at EACL26) we fix that. 1153
Reposted by Jan NiehuesAI4LT @ai4lt.bsky.social · 09/12/2025Last week @ecekt.bsky.social gave a great talk on her recent work "Model Merging to Maintain Language-Only Performance in Developmentally Plausible Multimodal Models"! Thank you so much for joining us, Ece! 🌟 011
Reposted by Jan NiehuesKIT-Fakultät für Informatik @kitinformatik.bsky.social · 06/12/2025聖誕老人究竟如何理解來自世界各地的眾多願望清單上的內容呢？ फादर क्रिसमस वास्तव में दुनिया भर की कई इच्छा-सूचियों पर लिखा हुआ कैसे समझ सकते हैं? How can Santa actually understand what is written on the many wish lists from all over the world? Find out at: ai4lt.iar.kit.edu #kitinformatik #ai4lt #llm #machinelearning #adventskalender #advent 094
Reposted by Jan NiehuesGianluca Vico @gianlucavico.bsky.social · 10/11/2025We’re collecting crowd-sourced translations in Piedmontese and Neapolitan. 🎯 Goal: see how well LLMs understand these languages. 👉 Participate here (in IT🇮🇹): - Piedmontese: quest.ms.mff.cuni.cz/crowd-transl... - Neapolitan: quest.ms.mff.cuni.cz/crowd-transl... Anyone can join, no need to be fluent!quest.ms.mff.cuni.czWelcome to CrowdTranslation 052
Reposted by Jan Niehuesmaikezufle.bsky.social @maikezufle.bsky.social · 10/11/2025It was great seeing you all at EMNLP in Suzhou. 🤗🌆 I presented two papers, check them out if you are interested in 🎤 Speech (summarization) or 📊 MT evaluation with multiple candidates. aclanthology.org/2025.emnlp-m... aclanthology.org/2025.wmt-1.63/ 062
Reposted by Jan Niehuesmaikezufle.bsky.social @maikezufle.bsky.social · 03/11/2025I am heading to Suzhou 🌆 for EMNLP, presenting two papers with Fabian Retkowski, @zouharvi.bsky.social and Tu Anh Dinh. Looking forward to see you at our posters about 🎤 Speech Summarisation (Fri. Nov 7, 14:00-15:30) and 📊 Machine Translation Evaluation (Sat. Nov 8, 11:00-12:00). 061
Reposted by Jan NiehuesAI4LT @ai4lt.bsky.social · 31/10/2025Excited 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 044
Reposted by Jan NiehuesAI4LT @ai4lt.bsky.social · 23/10/2025Wir 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👇 111
Reposted by Jan NiehuesKIT-Fakultät für Informatik @kitinformatik.bsky.social · 23/10/2025Am 30.10.25 lädt die VHS Wörth zu einem spannenden Abend rund um KI und Sprache ein. Maike Züfle, Doktorandin in der Forschungsgruppe Artificial Intelligence for Language Technologies am @kit.edu erklärt, wie große Sprachmodelle Texte verstehen, schreiben und übersetzen u. zeigt praktische Beispiele 012
Reposted by Jan NiehuesAI4LT @ai4lt.bsky.social · 25/08/2025🎉 Starting the week with some great news: 3 of our papers have been accepted to #EMNLP Main! 🧵👇 Big congratulations to all co-authors! 📚✨ Looking forward to presenting in #Suzhou! 132
Reposted by Jan NiehuesAI4LT @ai4lt.bsky.social · 18/08/2025Our team is growing! Today, we’re excited to welcome our new colleague Yuka Ko, who joins us as a postdoc on the EU Project DVPS. Looking forward to collaborating and achieving great things together. Welcome aboard! 🚀 041
Reposted by Jan NiehuesAI4LT @ai4lt.bsky.social · 18/08/2025#Interspeech2025 Excited to present our work on in-context language learning for endangered languages! 🤩 Come chat with Zhaolin Li today 14:30–16:30 (A09-P1-04)! 🔗 www.isca-archive.org/interspeech_... 041
Reposted by Jan Niehuesarxiv cs.CL @arxiv-cs-cl.bsky.social · 29/07/2025Sara 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 arxiv.org/abs/2507.19634 044
Jan Niehues @jan-niehues.bsky.social · 27/07/2025🚆 On my way to ACL 2025 in Vienna! AI4LT (KIT) will present work at ACL, SRW, Gem Workshop, and IWSLT. If you're attending and would like to chat about our research, feel free to reach out or drop by one of the sessions below 🧵. Looking forward to see you there 100
Reposted by Jan NiehuesAI4LT @ai4lt.bsky.social · 27/07/2025Excited to share 8 papers at #ACL2025 and affiliated venues! 🎉 Our team explored topics including multimodality, cross-lingual transfer, evaluation, decoding, and low-resource languages. Looking forward to presenting our work and connecting with the community! See you soon! 🤩 121
Reposted by Jan Niehuesmaikezufle.bsky.social @maikezufle.bsky.social · 26/07/2025🇦🇹 I’ll be in Vienna for #ACL2025NLP! Interested in training a speech-aware LLM without a lot of parameteres or data? Come to my poster: 🖼️ Mon, 18:00 Also into Speech Summarization? Join my IWSLT talk in collab with @fbk-mt.bsky.social 🎤 Fri, 14:00 Happy to chat - come say hi! 😎 Papers in 🧵 132
Reposted by Jan NiehuesAI4LT @ai4lt.bsky.social · 01/07/2025🚀 Today marks the official start of the #HorizonEurope DVPS project (dvps.translated.com). Happy that @kit.edu is part of this collaboration with 20 leading AI organizations! 📍 This week we're in Rome for the kickoff meeting. Looking forward to exciting collaborations! #DVPS #KIT #MultimodalAIdvps.translated.comHomepage - DVPS 031
Reposted by Jan NiehuesAI4LT @ai4lt.bsky.social · 28/05/2025Excited to share 7 papers accepted at #ACL2025 and affiliated venues! 🎉 Topics include multimodality, cross-lingual transfer, benchmarks, decoding, low-resource languages and more. See you in Vienna! 🤩 #nlproc #acl2025nlp 032
Reposted by Jan NiehuesAI4LT @ai4lt.bsky.social · 17/05/2025Wir freuen uns, bei der diesjährigen #TagDerOffenenTür @kit.edu dabei zu sein! Kommt vorbei und lernt #KITCat, unseren interaktiven KI-Assistenten kennen, und erlebt unser mehrsprachiges Sprachübersetzungssystem in Aktion! 053
Reposted by Jan NiehuesKIT-Fakultät für Informatik @kitinformatik.bsky.social · 12/05/2025Tag der offenen Tür mit Campus-Tag am #KIT am 17. Mai 2025, und wir sind auch dabei mit umfangreichem Programm: www.informatik.kit.edu/14708.php #TdoT 023
Reposted by Jan NiehuesAI4LT @ai4lt.bsky.social · 28/04/2025Also proud of our collaboration with @juliuscheng.bsky.social and @zouharvi.bsky.social on Bayesian optimization for MT reranking! 🤝 aclanthology.org/2025.naacl-l... 031
Reposted by Jan NiehuesAI4LT @ai4lt.bsky.social · 28/04/2025Excited to attend #NAACL in Albuquerque! 🎈🌞 We are presenting works on: 🌐 multilingual & multimodal representations in foundation models aclanthology.org/2025.naacl-s... 🔊 contextual learning for low-resource ASR (LM4UC workshop) aclanthology.org/2025.lm4uc-1.1 121
Reposted by Jan NiehuesAI4LT @ai4lt.bsky.social · 15/04/2025We had a great time hosting enthusiastic high schoolers at #GirlsDay2025! We explored #LLM basics, played "Break the AI" games, and shared study abroad opportunities at #KIT. Thanks @maikezufle.bsky.social for organizing and @kit.edu @girls-day.de for this initiative! 👩💻🌟 022
Reposted by Jan NiehuesIWSLT @iwslt.bsky.social · 03/04/2025The evaluation period has begun for our shared tasks! The test data is now available on our website, and submissions are due Tuesday April 15! ⏰ Please email task organizers or the google group with any questions 🥳 064
Jan Niehues @jan-niehues.bsky.social · 02/04/2025Just returning from Hannover Messe #HM25, where we had the chance to showcase the KIT Lecture Translator at the Baden-Württemberg Evening Reception. 191
Reposted by Jan Niehuessarapapi.bsky.social @sarapapi.bsky.social · 01/04/2025📢 The evaluation period of the Instruction Following task at @iwslt.bsky.social just started! 🖥️ Consider submitting your speech-to-text system! The outputs can be easily uploaded on the SPEECHM platform developed in the Meetween project (www.meetween.eu)! ➡️ iwslt2025.speechm.cloud.cyfronet.pliwslt2025.speechm.cloud.cyfronet.pl 085
Reposted by Jan NiehuesIWSLT @iwslt.bsky.social · 05/02/2025Why participate? 🔓 Help make the world’s vast audiovisual heritage accessible to all! 🔍 Join the scientific effort to push forward multilingual subtitling and improve global communication! 🌎 #SpeechTranslation #AIforGood 021
Reposted by Jan NiehuesIWSLT @iwslt.bsky.social · 05/02/2025🆕 For this year, Arabic joins the challenge! ✨ For the first time, we ask participants to generate subtitles for Arabic, covering the 5th most spoken language and one of the six UN 🌐 official languages. Let’s break language barriers! 🌍 #Subtitling #IWSLT2025 131
Reposted by Jan NiehuesIWSLT @iwslt.bsky.social · 05/02/2025Our last task: Subtitling!! 🎯 Goal: The IWSLT 2025 Subtitling Track challenges participants to generate accurate Arabic & German subtitles for English audiovisual recordings, bridging language gaps in media! 🌍📺 #AI #SpeechTech #Subtitling #IWSLT2025 🔗: iwslt.org/2025/subtitl...iwslt.orgSubtitling trackHome of the IWSLT conference and SIGSLT. 142
Reposted by Jan NiehuesIWSLT @iwslt.bsky.social · 04/02/2025Why participate? ✨ Contribute to advancing real-world systems for multilingual speech translation, particularly for underrepresented languages ✨ Collaborate to bridge linguistic divides & make this technology more accessible to millions of speakers worldwide 011
Reposted by Jan NiehuesIWSLT @iwslt.bsky.social · 04/02/2025For the 2nd iteration... 🆕 Significantly more data for low-resource Indic languages + more language directions! ➡️ Both English to Indic and Indic to English translation directions, addressing the unique challenges posed by low-resource linguistic contexts. 111
Reposted by Jan NiehuesIWSLT @iwslt.bsky.social · 04/02/2025Our final task highlight: Indic speech translation! 🎯 Goal: develop a speech-to-text translation model that bridges the gap for low-resource Indian languages, focusing on Hindi, Bengali, and Tamil from the Indo-Aryan and Dravidian families. 🔗: iwslt.org/2025/indiciwslt.orgIndic trackHome of the IWSLT conference and SIGSLT. 121
Reposted by Jan NiehuesIWSLT @iwslt.bsky.social · 03/02/2025Why participate? ✨Low-resource is truly the frontier, and there are many ways to making an impact for a language community: from providing a dataset for a language that is missing one, to exploring the limits of different models and techniques. 011
Reposted by Jan NiehuesIWSLT @iwslt.bsky.social · 03/02/2025🆕 This year includes new and more diverse language pairs (e.g. Fongbe to French, Estonian to English) in addition to continuing pairs 🆕🔊 Also new this year is a *data track* which encourages the creation of new speech translation datasets for less-supported languages 111
Reposted by Jan NiehuesIWSLT @iwslt.bsky.social · 03/02/2025Today's task: Low-resource ST! 🎯 Goal: Building speech translation models for currently underserved, mostly low-resource languages and varieties 🗓️ This is the 5th iteration, with new and continuing language pairs (10 total!) 🔗: iwslt.org/2025/low-res...iwslt.orgLow-resource trackHome of the IWSLT conference and SIGSLT. 121
Reposted by Jan NiehuesIWSLT @iwslt.bsky.social · 31/01/2025Why participate? 🚀 - An #IWSLT tradition: the perfect arena to set the state-of-the-art in Spoken Language Translation, test new solutions, join an active research community & push AI’s limits to transform global communication! 🌍🎙️🤖 021
Reposted by Jan NiehuesIWSLT @iwslt.bsky.social · 31/01/2025What is new and different this year? ✨🆕 – The challenge gets tougher! Participants will engage with Arabic as a new target language (alongside German & Chinese) 🌍 and diverse testing scenarios featuring unique speech styles, accents & recording conditions. 🎤 Are you ready? 111
