{"id":"arxiv/2610.24323","url":"https://arxiv.org/abs/2610.24323","tags":["math.OC","cs.GT"],"score":0,"title":"Computing Multi-Slot First-Price Pacing Equilibria at Scale via Entropic Smoothing","domain":"math.OC","source":"arxiv","authors":["Yuchen Xie","Shenyinying Tu","Yuan Gao","Yi Zhang","Deepak Kumar"],"reviews":0,"subject":"math.OC","summary":"Pacing equilibria describe how budget-constrained advertisers clear an ad market, but computing one at production scale is hard when every request fills several slots: the allocation jumps whenever two campaigns' scores cross. We consider the first-price pacing equilibrium (FPPE) of auctions with $K$ homogeneous slots and unit-demand campaigns, whose Eisenberg-Gale dual reduces, via a TopK identity, to a convex program over bids. Regularizing each request's allocation with the binary (Fermi-Dirac) entropy gives a smooth, strictly convex dual whose gradient is exactly spend minus budget and…","category":"research","comments":0,"disputed":false,"division":"math","mentions":0,"discussed":0,"published":"2026-10-16","reactions":{}}