Co-author here. This paper proposes random bridges as an alternative transport mechanism for generative models. The core idea is that instead of using diffusion processes (which require many sampling steps and are computationally expensive), we use stochastic bridge processes to transport between distributions.
The key results:
Competitive FID scores with significantly fewer sampling steps compared to standard diffusion models
The framework is flexible — it generalises several existing approaches including score-based diffusion and flow matching
Reduced computational cost at inference time while maintaining sample quality
We developed this at A.I.M. Research Lab with co-authors from UCL and the Bank of England. The motivation came from practical experience with generative models in production settings where inference cost matters.
Paper: https://arxiv.org/abs/2512.14190
Happy to answer questions about the theory or practical implications.
Co-author here. This paper proposes random bridges as an alternative transport mechanism for generative models. The core idea is that instead of using diffusion processes (which require many sampling steps and are computationally expensive), we use stochastic bridge processes to transport between distributions. The key results:
Competitive FID scores with significantly fewer sampling steps compared to standard diffusion models The framework is flexible — it generalises several existing approaches including score-based diffusion and flow matching Reduced computational cost at inference time while maintaining sample quality
We developed this at A.I.M. Research Lab with co-authors from UCL and the Bank of England. The motivation came from practical experience with generative models in production settings where inference cost matters. Paper: https://arxiv.org/abs/2512.14190 Happy to answer questions about the theory or practical implications.