David Yan
@david-yan.bsky.social
undergrad @ princeton doing computer vision stuff
https://david-yan1.github.io
https://david-yan1.github.io
Another good example from mono depth: DepthAnythingV2 uses a teacher supervised only on synthetic data (600k) and students are distilled from its predictions on web images (62M).
Real-world GT is noisy, so fitting to limited, but perfect synthetic data is better for teacher accuracy.
Real-world GT is noisy, so fitting to limited, but perfect synthetic data is better for teacher accuracy.
July 16, 2025 at 9:57 PM
Another good example from mono depth: DepthAnythingV2 uses a teacher supervised only on synthetic data (600k) and students are distilled from its predictions on web images (62M).
Real-world GT is noisy, so fitting to limited, but perfect synthetic data is better for teacher accuracy.
Real-world GT is noisy, so fitting to limited, but perfect synthetic data is better for teacher accuracy.
Threw it at o3 and after thinking for 12 min (!!) it gave “O Canada” and A major. You can take a look at the full chain of thought here (chatgpt.com/share/6871b3...). Some highlights:
July 12, 2025 at 1:03 AM
Threw it at o3 and after thinking for 12 min (!!) it gave “O Canada” and A major. You can take a look at the full chain of thought here (chatgpt.com/share/6871b3...). Some highlights: