Channel: #geo-regional-asia
About:
Geo-Regional Asia is where researchers from across Asia come together to share what they're working on and learn from each other. We've had the privilege of hosting over 75 research sessions, and it's been incredible to see the diverse range of work coming out of our region.
We also love bringing people together for deeper learning experiences through small cohort programs. Recently, we gathered a group to work through NYU's Deep Learning course taught by Yann LeCun - it was amazing to see everyone supporting each other through the challenging material and building lasting connections along the way.
At its heart, this community is about celebrating the brilliant work happening across Asia and making sure these voices are heard. Whether you're presenting your research or just curious to learn, you'll find a welcoming space here to connect with fellow researchers who share your passion.
Co-leads:
Ahmad Mustafa Anis - .ahmadmustafaanis on Discord, @AhmadMustafaAn1 on Twitter
Kanwal Mehreen - .kanwal_mehreen on Discord, @KanwalMehreen2 on Twitter
Recent Presentations
Xiuyu Li presents Q-Diffusion: Quantizing Diffusion Models
On the reliability of watermarks for LLMs with Khalid Saifullah
Adaptation to Distribution Shifts in an Unsupervised Manner with Jehanzeb Mirza
LLMatic: Neural Architecture Search via Large Language Models and Quality-Diversity Optimization Paper
Generating Images with Multimodal LMs with Jing Yu Koh
Adversarial Attacks on Aligned LLMs by Andy Zou
Fabio Cermelli: CoMFormer
Sanjay Subramanian, Visual Reasoning with Limited Human Labels
Badr AlKhamissi presents OPT-R: Exploring the Role of Explanations in Finetuning and Prompting for Reasoning Skills of Large Language Models
Geometry and topology in memory and navigation, Tom Burns
Demonstrate-Search-Predict with Omar Khattab
Code-Assisted Reasoning with LLMs with Aman Madaan
DEtection TRansforme (DETR) prestented by @tahreem
Pathways to semi-unsupervised NLP brain, with Muhammad Saiful Bari
Nedjma Ousidhoum presents "What is Needed Vs. What is Built in NLP: Toxic Language Detection and Automated Fact-checking Models as Use Cases"
Segment Anything Model (SAM) by MetaAI
Zubair Irshad - Learning object-centric 3D scene representations
Ahmad Anis presents "SimCLRv2" paper by GoogleAI.