KAUST Computer Vision Lab (IVUL) (@KaustVision) header
KAUST Computer Vision Lab (IVUL) (@KaustVision) profile picture

Image and Video Understanding Lab (IVUL) @KAUST_News supervised by Prof. @BernardSGhanem

KAUST, Saudi Arabia
Joined March 2020
Photo by KAUST Computer Vision Lab (IVUL) (@KaustVision)Photo by KAUST Computer Vision Lab (IVUL) (@KaustVision)Photo by KAUST Computer Vision Lab (IVUL) (@KaustVision)Photo by KAUST Computer Vision Lab (IVUL) (@KaustVision)Photo by KAUST Computer Vision Lab (IVUL) (@KaustVision)Photo by KAUST Computer Vision Lab (IVUL) (@KaustVision)Photo by KAUST Computer Vision Lab (IVUL) (@KaustVision)Photo by KAUST Computer Vision Lab (IVUL) (@KaustVision)Photo by KAUST Computer Vision Lab (IVUL) (@KaustVision)Photo by KAUST Computer Vision Lab (IVUL) (@KaustVision)
KAUST Computer Vision Lab (IVUL) retweeted
Our paper, "Multimodal Safety Evaluation in Generative Agent Social Simulations," has been accepted to #ACL2026NLP! 📰Stop by Adonai's presentation today, July 5, in the Grand Hall from 4:00–5:30 PM (Paper #5266) @AdonaiVera19, @CarlosH_93, H Hamid, D Kim, @BernardSGhanem
1
4
11
837
KAUST Computer Vision Lab (IVUL) retweeted
Frontier labs spend millions purchasing RL environments for training terminal agents. But we decided to open source it. Introducing SETA: Scaling Environments for Terminal Agents, the largest open source training RL environments for terminal agents. We released: - 400 termianl agent training environments, more to come - SOTA agent harness on terminal-bench with CAMEL terminal toolkit - The RL training pipeline and trained SETA-RL-Qwen3-8B model weights
26
91
6
770
227,170
KAUST Computer Vision Lab (IVUL) retweeted
🧵1/N 🔥Introducing EasyV2V from Snap Research! 📌TLDR: We propose a simple and powerful model for instructional video editing! Just use a text prompt to edit your video! #Aiart #AIArtwork #Prompt #ArtificialIntelligence #GenerativeAI #MachineLearning #AI
7
5
2
15
4,877
KAUST Computer Vision Lab (IVUL) retweeted
Wenxuan Zhu (@WenxuanZhu75798) will present our ICCV paper today: “4D-Bench: Benchmarking Multi-modal Large Language Models for 4D Object Understanding”, co-authored with @1zhengcheng , Chiawen Lin, Jinjie Mai, Chen Jun, @Eng_Hemdi ,@SarisRo , @moElhoseiny ,@BernardSGhanem 😊We introduce the first comprehensive benchmark evaluating MLLMs’ ability to understand dynamic 4D objects. 🎃 Our findings reveal a critical gap: even state-of-the-art MLLMs struggle with 4D object understanding. #iccv25 #ICCV2025 @KaustVision webpage: wenxuanzhu1103.github.io/4db… paper: arxiv.org/pdf/2503.17827
2
1
468
KAUST Computer Vision Lab (IVUL) retweeted
Thank you very much, Prof. @Yoshua_Bengio, for accepting our invitation. It was a great honor to have you at the First edition of the @SaFeMMAI Workshop on Safe and Trustworthy Multimodal AI Systems. 🛡️Talk: "Avoiding Catastrophic Risks From Uncontrolled AI Agency" #ICCV2025
2
9
670
KAUST Computer Vision Lab (IVUL) retweeted
Our afternoon session at the #SafeMMAI Workshop @ICCVConference is starting soon! ☀️ Join us for more exciting discussions, and don’t forget to grab your KAUST USB swag! 🎁 #ICCV2025 #KAUST #AI
1
1
6
363
KAUST Computer Vision Lab (IVUL) retweeted
👇 Check out our full schedule
1
3
272
KAUST Computer Vision Lab (IVUL) retweeted
Heading towards Hawaii 🌺✈️ Excited to present MatchDiffusion at #ICCV2025! 🗓️ Wed 22 Oct, 2:45–4:45 p.m. HST 📍 Exhibit Hall I, #1373 📄 Paper: arxiv.org/abs/2411.18677 Poster info: iccv.thecvf.com/virtual/2025… #GenerativeAI #MatchCut #ComputerVision See you there!!
2
1
7
686
KAUST Computer Vision Lab (IVUL) retweeted
Ready to rise among the next generation of #AI innovators? Apply for the #KAUST Rising Stars in AI Symposium, organized by the #Center of Excellence for Generative AI, and join a global cohort of AI innovators! Deadline: Nov 15, 2025 🔗 kaust.edu.sa/events/rsais26
1
11
2
20
2,120
KAUST Computer Vision Lab (IVUL) retweeted
#KAUST welcomes emerging global talent to the Rising Stars in AI Symposium 2025. President Sir Edward Byrne AC: “You are all incredibly lucky to be working in the most important field in the world today — a field that is pivotal for our age and for our success as a planet.”
1
5
25
2,595
KAUST Computer Vision Lab (IVUL) retweeted
Rising stars for global impact. Hosted by the Center of Excellence for Generative #AI, the “Rising Stars in AI Symposium 2025” at #KAUST brought together the next generation of talent shaping the future of AI research and innovation. Read more: kaust.news/XXmO
2
3
1
34
3,865
KAUST Computer Vision Lab (IVUL) retweeted
Excited to share our new paper "Beyond the Last Answer: Your Reasoning Trace Uncovers More than You Think"! We show that digging into an LLM’s intermediate “subthoughts” and aggregating their answers can significantly boost math reasoning performance. A thread 🧵
4
9
4
22
4,808
KAUST Computer Vision Lab (IVUL) retweeted
Go check our poster! @MereyRamazanova will be there and will explain: How to deal with a pretrained network with missing modalities at test time? 💡 Don’t miss it!
🌟We are excited to be presenting two papers at @iclr_conf ICLR 2025 in Singapore! 🙌🇸🇬 Catch us there from April 24 to 28! 👇Check out the details of both papers in the image below. #ICLR2025
1
3
494
KAUST Computer Vision Lab (IVUL) retweeted
Just released "DiffCLIP", extending Differential Attention proposed by @ytz2024 to CLIP models - replacing both visual & text encoder attention with the differential attention mechanism! TL;DR: Consistent improvements across all tasks with only 0.003% extra parameters!
5
10
27
24,119
KAUST Computer Vision Lab (IVUL) retweeted
Convex splatting accepted at @CVPR #CVPR2025
📢Happy to present Convex Splatting, a novel way for 3D reconstruction based on 3D smooth convexes. For the first time, a splatting-based method reaches the quality of NeRF sota methods but with real-time rendering and few primitives!! I expect this to replace Gaussian splatting for 3D in the coming months. CODE RELEASED TODAY! joint work with collaborators from @UniversiteLiege @Oxford_VGG , @KaustVision a thread 🧵 1/n
3
9
106
9,201
KAUST Computer Vision Lab (IVUL) retweeted
LIVE NOW! ⚽🔥 📽️ Check out the official #SoccerNet 2025 video! Find all the details about this year's competition here 🏆. 📅 Two months for the deadline: April 24, 2025 🏅 Award Ceremony: @CVPR 2025 🔁 Repost & spread the word! 📺 Watch now: youtu.be/J3c4YTME5ZQ
3
8
2
17
21,569
إلى من طاردوا المستحيل ولم يخافوه، وأرهقهم السؤال ولم يتجاهلوه! شكرًا لأنكم خضتم غمار المجهول وأعدتم تعريف الممكن.. عقول مبدعة تنال عن جدارة جائزة التميز البحثي في اقتصاديات المستقبل.
1
10
2
23
8,862
استكشف الخبراء عبر الجلسات الحوارية ضمن فعالية الابتكار في اقتصاديّات المستقبل الدور المحوري للذكاء الاصطناعي في اتخاذ القرارات الفاعلة، ومساهمة البنية التحتية البحثية بالمملكة في إطلاق قدراته والاستفادة منه في إيجاد حلول نوعية للعديد من التحديات المحلية والدولية.. للمشاهدة : youtu.be/UG-qxcMFVjs?si=4_Du… youtu.be/7c8v5wGzc2I?si=5Urv…
5
1
8
5,123
KAUST Computer Vision Lab (IVUL) retweeted
🎉Thrilled to share that our paper, "Test-Time Adaptation for Combating Missing Modalities in Egocentric Videos", has been accepted to #ICLR2025! Grateful to my amazing collaborators @PardoAlejo @BernardSGhanem @MotassimF 📜Paper: openreview.net/forum?id=1L52… 🌏See you in Singapore!
1
3
11
906