The fifth article in our ELI5 series is on Metric Learning by Keerat Kaur Guliani medium.com/vlgiitr/metric… Hope you enjoy the animations and explanation! #DeepLearning #MachineLearning #Mathematics

Join us for a discussion on Sunday(21/05/23), where we'll be discussing an introduction to generative AI Date:21st May Time:6:30pm Venue:ACM Room 109,SAC Topic: An Introduction to Generative AI No prerequisites required. See you All!!

We are releasing Open Projects in various AI domains. Submission Deadline: 22 June For the link to the google form and for more information regarding each project, visit our website or check the link in instagram bio. instagram.com/p/CseSBkOPhM3/…

Join us for a discussion, where we'll be discussing an introduction to reinforcement learning. Date: 24th May Time: 6:30pm Venue: ACM Room 109,SAC Topic: Reinforcement Learning No prerequisites required.

Join us as we explore the advanced ML/DL techniques and the mathematical background behind Recommender Systems. Date: 31st May Time: 6:30 pm Venue: ACM Room 109, SAC Topic: Recommender Systems No prerequisites required. See you all! instagram.com/p/Cs4SHnihqrA/…

📢 Introducing our #ECCV2024 work, COCO-ReM (COCO Refined Masks), for more reliable benchmarking of object detectors, crucial for the future of object detection research. Paper: arxiv.org/abs/2403.18819 Code: arxiv.org/abs/2403.18819 Website: cocorem.xyz

LLM post-training requires a reward model for preference alignment. But is it necessary? In our new preprint, we show that the language model is itself a preference classifier & reward model isn't needed. We can self-improve a model by using its own feedback!

New ICLR 2025 blogpost shows that a vision transformer with 2D ALiBi achieves the best performance extrapolating to larger image sizes on Imagenet100. Cool! iclr-blogposts.github.io/2025/blog/posi…

Excited to announce that my paper has been accepted in ACL main conference! The first one from Lossfunk - many more to come! 🧵 on the paper