OneBid improves advertising bids across multiple cost goals
OneBid: A Unified Auto-Bidding Foundation Model for Diverse oCPX Advertising Scenarios
Machine LearningArtificial Intelligence
Summary
Bidding for online ads needs to balance cost and getting valuable actions like signups or purchases. The paper presents OneBid, a single machine learning model that can work across different ad goals, replacing multiple specialized systems. It learns from past bidding logs and adapts safely without risky online experiments. Tests at a large platform showed it improved overall ad value by 2.2% and up to 13.1% in some cases.
What this means in practice
- •For digital advertising teams: Use OneBid to unify bidding models for multiple campaign goals, improving cost-effectiveness and simplifying pipelines.
- •For mobile game marketers: Enhance in-game ad campaign performance by controlling bids across varied user actions with a single adaptive model.
Authors
Yewen Li, Peng Jiang, Yitian Li, Pengfei Lv, Xialong Liu, Peng Jiang, Qingpeng Cai
Abstract
Auto-bidding is central to computational advertising, where strategies must maximize advertisers' conversion value under economic constraints. It has evolved from rule-based controllers to reinforcement learning and generative methods such as Decision Transformer (DT). Yet these methods increasingly mismatch the prevailing optimized cost-per-X (oCPX) paradigm, which spans heterogeneous scenarios (e.g., registration, purchase), each served by a separate model, leading to fragmented pipelines and underexploring cross-scenario modeling. Inspired by foundation models like LLMs, unifying these oCPX scenarios into one model raises three challenges: multi-objective control, scalable capacity under strict latency, and safe offline policy improvement. We present OneBid, a unified auto-bidding foundation model that learns a reusable backbone from heterogeneous oCPX logs and adapts it to scenario-specific deployments via offline post-training. Building on DT, OneBid extends single Return-to-Go conditioning to two atomic signals, Return-to-Go for conversion value and Cost-to-Go for cost ratio, plus value-aware regularization on next-action prediction. To absorb distributional heterogeneity, we design a sequence-level Mixture-of-Experts architecture, where shared experts encode cross-scenario knowledge and sparsely-routed experts capture scenario-specific patterns at low latency, yielding consistent scaling with model size and data. During post-training, we align the backbone with scenario preferences via Critic-guided Relative Offline Policy optimization (CROP): a learned critic scores candidate actions group-relatively, avoiding the unsafe online exploration of GRPO-style fine-tuning while constraining policy shift to reduce OOD risk. Validated via online A/B tests and fully deployed at Kuaishou, OneBid delivers an overall +2.2% ADVV gain on oCPX Ads, peaking at +13.1% in the ROAS scenario.