The authors introduce Painter-Thinker (PaTh), an architecture that separates reasoning from image generation.
Recursive latent reasoner boosts diffusion models for visual puzzles
PaTh solves 92.5% of hard MNIST Sudoku and 71.2% of extreme instances, far surpassing prior methods with far fewer parameters.
Top University
Paweł Skierś · Małgorzata Grzanka · Wojciech Masarczyk · Jan-Willem van de Meent · Kamil Deja
Warsaw University of Technology · IDEAS Research Institute · University of Amsterdam
Research Digest··3 min read
The authors propose Painter-Thinker (PaTh), which couples a small recursive reasoner (the Thinker) to a frozen diffusion model (the Painter) via ControlNet adapters.
Why this paper
From University of Amsterdam and 2 others
In one line
A recursive reasoner coupled to a frozen diffusion model solves visual reasoning puzzles from pixels without symbolic supervision.
What we could check
- ·No code link found
- ·No weights link found
- ·No dataset link found
- ✓Compute or model size stated (params 10M)
- ·No stated limitations found
- ✓Reports numbers on named benchmarks
Observed from the paper text and links we have. Absence here means we did not find it, not that it does not exist.
§