5/ I loved those passages. Even before we know who we’ll become or what we might accomplish, there are things we can do right where we are. While we wait for the future to become clearer, our lives can still be taking root.
4/ As a young man, Liu Bei chooses to return home to his mother instead of studying with a scholar his teacher has recommended. He has no clear plan yet, but wants to believe that putting down roots is the right place to begin.
Back home, he builds relationships and lays the foundations of his future.
I also paused at a passage about Cao Cao, another central figure. His quiet years back home are described as a time that prepared him for what lay ahead.
3/ I’ve been rereading Romance of the Three Kingdoms, a classic Chinese historical novel set in an age of war and shifting power. A few scenes have stayed with me.
What a fun week at #COLM2026! Great people, inspiring research, and a chance to meet @AndrewYNg.
Loved @chelseabfinn’s keynote! Also enjoyed several VLM papers on egocentric understanding & #spatialintelligence—so relevant to my work on #AR delivery glasses at @AmazonScience.
@AmazonScience@COLM_conf Thanks @AmazonScience team! I really enjoyed sharing my journey from Amazon’s Astro robot to AR delivery glasses at @COLM_conf —from multimodal research to building, testing, and deploying AI in the real world.
AI-powered smart glasses that guide delivery drivers from van to the customer's doorstep, hands-free. Hear from Amazon applied scientist @_yelinkim on the computer vision and edge AI behind it at @COLM_conf. #COLM2026
Hi, I’m Yelin! 👋 I lead CV/ML science for Amazon’s last-mile delivery glasses, working on edge AI and spatial intelligence.
I’m at #COLM2026 in SF this week—would love to grab coffee with others working on robotics, AR, or multimodal AI!
RSP achieves SOTA on EgoIntention and RIO. I’m especially excited about grounding user intent from a first-person view. #egocentric#augmentedreality#arglass
Our RSP paper is heading to #NeurIPS2026! ���
RSP connects language reasoning with precise object detection by turning MLLM attention into learnable spatial priors.
Proud of this work led by @mcagrigungor during his internship at @AmazonScience!
@AmazonScience Thanks for having me, @AmazonScience! My team is actively hiring Applied Scientist Interns and Postdocs in edge AI and SLAM, so drop by my Amazon booth session at #cvpr2026 today at 2 PM or DM me to connect!
Meet Yelin Kim – she leads computer vision for Amazon's Last Mile team, relocating AI from the cloud to wearables and enabling hands-free delivery with multimodal sensing.
Hear more about her work and check out her paper that was selected as a #CVPR2026 highlight: https://t.co/5zSNNsq6nC