📢 Excited to announce our new study, "Evaluating Linguistic Capabilities of Multimodal LLMs in the Lens of Few-Shot Learning"! 🚀
- Paper: https://t.co/Sx0xUOXYsV
#AI#NLProc (1/6)
(iii) Our findings highlight the importance of pretraining data composition and the potential of few-shot learning strategies to improve the reasoning abilities of MLLMs. (5/6)
I'm thrilled to announce that I have successfully defended my Master's thesis! I am deeply thankful to my advisors, Erkut Erdem (@erkuterdem) and Aykut Erdem (@aykuterdemml) for their unwavering support and guidance. I thoroughly enjoyed our work together throughout this journey.
Congratulations to our student Mustafa Dogan (@im_mdogan) on successfully defending his Master's thesis last Friday! Special thanks to the jury members Nazli Ikizler Cinbis (@nazliikizler) and Sinan Kalkan (@kalkansinan) for their invaluable insights and contributions!
@fkadev Early return biz de yer yer kullanıyorduk ama statik analiz yaparken de bu sefer MISRA kurallarından kalıyorduk. Bütün early returnleri kaldırmam gerekmişti bu yüzden
Bu dönem Yapay Zeka Mühendisliği lisans programında verdiğim 'Yapay Öğrenmenin Temelleri' dersinde, ders projesinin temasını "Sürdürülebilirlik ve Yapay Öğrenme" olarak belirlemiştim.
📢Thrilled to share that our work "ViLMA: A Zero-Shot Benchmark for Linguistic and Temporal Grounding in Video-Language Models" is accepted to #ICLR2024! @iclr_conf#NLPro
- Paper: https://t.co/PxP3xQdobI
- Site: https://t.co/F36vJvnHui
- Code: https://t.co/mkGWiq3rrz
🎉 Exciting news! Our paper with Profs. Kasım Öztoprak & Mehmet Reşit Tolun on 'Teaching computer architecture by designing & simulating processors from their bits and bytes' got accepted in @PeerJCompSci! Two years of dedication paid off—a proud first journal publication! 📚💻
@DisneyPlus@ahsokaofficial You should be ashamed. You have disrespected a successful historical character who pioneered for his country and won the recognition of the whole world, who was modern, progressive and recognized women's rights before many European countries.
📢Excited to announce that our work on event-based vision has been accepted to CVPR 2023 Workshop on Event-Based Vision! Check out our proposed evaluation methodology and open-source framework EVREAL for comprehensive analysis of event-based video reconstruction methods #CVPR2023
Happy to share that our work on video generation from events got accepted to CVPR 2023 Workshop on Event-Based Vision. @ercanbur and Onur did an amazing job building a benchmark suite to assess the model performances in depth. Stay tuned for our follow-up works! #eventcameras