AI RESEARCH
A Differentiable Bayesian Relaxation for Latent Partial-Order Inference
arXiv CS.LG
•
ArXi:2605.06976v1 Announce Type: cross Many ranking and agent trace datasets are recorded as linear orders even though their latent structure is only partially ordered. This is especially common in agent and workflow traces, where observed order may reflect arbitrary linearization rather than true prerequisites. We