p

Join Mu-SHROOM 🍄, a SemEval 2025 shared task on detecting hallucination spans in multilingual LLM outputs! 🌍 Includes Czech with regional Czech questions 🇨🇿. Do you think you can spot when something isn’t true? 🤔 Try it out! 👉 helsinki-nlp.github.io/shroom #SemEval2025 #NLProc

Super excited for tomorrow's UniBw AI Retreat! 💡 Looking forward to inspiring talks, hands-on tutorials and workshops. Let’s build and grow the UniBw AI network together! 🚀 We still have three days left to see the mountains 🏔️😅 @unibw_m @HSUHamburg @Marshall_Center

🚨 Join the UniBw AI Retreat! 🚨 📍 Marshall Center, Garmisch 📅 10-12 Dec 2024 ✨ Talks on AI & security 🛠 PyTorch tutorial 🔍 Open science workshop 🎨 Poster presentations ⏳ Apply by 18 Nov 2024: go.unibw.de/ubair2024 For @unibw_m, @HSUHamburg and security authorities!

👏 A big thank you to @MayBritt_NFG, who organized the event on behalf of the Marshall Center. Also a big thank you to @RafaelaKraus and @dtecbw for the financial support.

🚀 Join the UniBw AI Retreat at @Marshall_Center! ▶️ Showcase your Research ▶️ Adv. PyTorch Tutorial ▶️ Open Science Workshop ▶️ Top Speakers on AI & Global Security 📢 Open to @unibw_m, @HSUHamburg & Public Sector AI Researchers 👉 Register now: go.unibw.de/ubair2024

Explore our new dataset for detecting tiny changes in images or text for vision-language tasks! Ideal for benchmarking multimodal models requiring high precision in fine-grained variations. #ACL #VisionLanguage #DatasetRelease @unibw_m @ufal_cuni

Enhancing Conceptual Understanding in Multimodal Contrastive Learning through Hard Negative Samples at the ALVR workshop. aclanthology.org/2024.alvr-1.9 Presented by @jlibovicky with @phiyodr and folks from @unibw_m.

@NitzlChristian Do you already have a link to the paper?

🐶📢Only two days left until our "1st Workshop on Vision-Based Structural Inspections in Civil Engineering" @ #WACV2024 on Sunday, January 7th! The room is Paniolo I-II. Attached you can find the timetable. We are looking forward to welcoming you!

▶️The eval_ai team has fixed the database error and the evaluation is running again. Due to that we will #extend the submission period. Both, End of Challenge and Fact-Sheet Submission will be on November 14, 2023 (23:59:59 UTC)! #daclchallenge #wacv2024 eval.ai/web/challenges…

📢The data split for the Testfinal Phase was released! It can be downloaded here: GigaMove: gigamove.rwth-aachen.de/en/download/38… or AWS: dacl10k.s3.eu-central-1.amazonaws.com/dacl10k-challe… We wish you good luck! Don't forget to upload the fact-sheet until November 12, 2023 (23:59:59 UTC)!

📢We have an on-going #SemanticSegmentation challenge @eval_ai for our workshop @wacv_official ▶️Domain: #CivilEngineering ▶️Task: Detect damages & other objects on bridges ▶️Dataset: #dacl10k with about 10k images 🏆Win up to $3.000 #evalai #WACV2024 eval.ai/web/challenges…

We have just integrated the #dacl10k dataset into #FiftyOne. With this handy visualization tool you can easily explore the dataset. Thanks to @Voxel51! Don't miss our dacl-challenge @wacv_official and win up to $3,000! try.fiftyone.ai/datasets/dacl1… dacl.ai/workshop #WACV2024

🎉We are happy to announce the 1st Workshop on Vision-Based Structural Inspections in Civil Engineering @wacv_official🌉⛑️ Be part of the dacl-challenge - a challenge for Semantic Bridge Damage Segmentation and wiz prizes up to $3.000🏆 @unibw_m @bauhaus_uni #WACV2024 #dacl10k

➡️Paper: arxiv.org/abs/2309.00460 ➡️Dataset: #arrow_right-data" target="_blank" rel="nofollow noopener">github.com/phiyodr/dacl10… ➡️Workshop: dacl.ai/workshop.html

🎉 We are happy to announce the first large semantic segmentation dataset for damage and object recognition on bridges! "dacl10k" comes with 10k images and 19 classes - all important for structural safety inspections!🌉⛑️ @unibw_m @dacl_ai #dacl10k ⤵️Details below

I will need another #PhD student 👩‍🎓👨‍🎓🧑‍🎓 starting in March 2024. If you would like to try out what it is like to work with me, try out @ufal_cuni's pre-PhD internship this fall.

@drfranksauer Unsere Studenten haben auch schon herausgefunden, dass man damit die Aufgaben aus "Höhere Programmierung" lösen kann 💻

📢#unibw_m/#hsu researchers: Join us for the 3rd #MLIG STAMMTISCH on 17th November 2022 at #datArena (reopened 🖥️💻⌨️💾 museum @unibw_m) and dinner/beer at Brandl :) ➡️unibw.de/vis/mlig #MachineLearningInterestGroup

Registration for tomorrow's #MLIG ONLINE MEETUP is still open. We will have talks about #ML/#AI in #CyberSecurity, #SocialScience, and #CivilEngineering 📢Internal event at @unibw_m and @HSUHamburg 💻 unibw.de/vis/mlig