Author

Daniel Csillag

1 paper indexed here

We haven’t gathered this author’s papers yet. Follow them and we’ll fetch their work.

Not the right person? Other researchers publish under this name.

#artificial intelligence Preprint Oct 2026

Rethinking GFlowNet Training: Negative Reinforce Dynamics

This paper establishes a connection between the full gradient and min-min dynamics between the GFlowNet and a decoupled sampler, showing that the sampler reduces observed violations by avoiding high-error regions.

Pedro Dall'Antonia, Daniel Csillag, Diego Mesquita · 0 citations