Vol. 01 · Vancouver, BC · Kept since 2026
Field Notes
of an AI Researcher
Aditya Chinchure
I am a PhD student at the Computer Vision and NLP Labs at University of British Columbia. My research focus is on applying methods from human reasoning and cognition to develop vision-language models and evaluation tools.
Reaction to Tesla’s Roadster
Wow, we got the flying car! It’s always impressive to see how Elon & team executes projects with sheer determination. It is clear that flying version of the roadster will likely never get past the demo vehicle (aka…
118 words · Oct 21
Injecting small VLMs with strategies for tool-use on videos
205 words · Oct 21
Testing the iPhone 18 Pro for photography
125 words · Oct 21
Spotlight: Identifying and Localizing Video Generation Errors Using VLMs
116 words · updated Jun 1
SPIKE-RL: Video-LLMs meet Bayesian Surprise
105 words · updated Apr 1
Black Swan: Abductive and Defeasible Video Reasoning in Unpredictable Events
240 words · updated Feb 1Sorted by last updatedAll 6 notes →
Ideas I'm working on now, updated as they evolveOpen notebook →
Spotlight: Identifying and Localizing Video Generation Errors Using VLMs
116 words · 1 minUpdated
Jun 1, 2026
Jun 1, 2026
SPIKE-RL: Video-LLMs meet Bayesian Surprise
105 words · 1 minUpdated
Apr 1, 2026
Apr 1, 2026
Black Swan: Abductive and Defeasible Video Reasoning in Unpredictable Events
240 words · 2 minUpdated
Feb 1, 2025
Feb 1, 2025
3 notesOpen notebook →