Conditional Generation of Creative Chess Puzzles with Diffusion Models
Abstract
A masked diffusion model with an auxiliary best-move prediction task and reinforcement learning via DDPO generates controllable, unique chess puzzles conditioned on tactical themes and partial board positions.
While modern language models demonstrate impressive generative capabilities, they often struggle with constrained, counter-intuitive creative tasks. To address this limitation, we explore chess puzzle generation as a rigorous testbed for computational creativity and reasoning, a domain where altering a single piece can invalidate an entire solution. We propose a novel approach for conditional generation of creative chess puzzles using masked diffusion models. Unlike previous methods, our non-directional diffusion approach allows for conditioning on specific tactical themes and partial board positions. We introduce a novel auxiliary task of simultaneous best-move prediction, which improves solution uniqueness by 11.6% and theme-conditioning accuracy by 2.5%. To further optimize solution uniqueness and theme conditioning, we establish a reinforcement learning framework adapted from Denoising Diffusion Policy Optimization (DDPO). This RL training increases the yield of unique and theme-matching positions by 89.1%. Finally, we release the first open-weights models (Appendix B) for chess puzzle generation, offering a new pathway for controllable, creative generation.
Models citing this paper 1
Datasets citing this paper 0
No dataset linking this paper
Spaces citing this paper 0
No Space linking this paper
Collections including this paper 0
No Collection including this paper