Unified Value Alignment for Generative Recommendation in Industrial Advertising

TL;DR

UniVA introduces a unified framework for value-aligned generative recommendation in industrial advertising, combining tokenization, decoding, and online serving to optimize both user interest and commercial value.

Contribution

It proposes a novel unified value alignment framework with a value-discriminative tokenizer, generation-as-ranking decoder, and value-guided beam search for improved advertising recommendation.

Findings

37.04% improvement in offline Hit Rate@100

1.5% GMV lift in online A/B tests

Effective alignment of value signals across system components

Abstract

Generative Recommendation (GR) reformulates recommendation as a next-token generation problem and has shown promise in industrial applications. However, extending GR to industrial advertising is non-trivial because the system must optimize not only user interest but also commercial value. Existing GR pipelines remain largely semantics-centric, making it difficult to align value signals across tokenization, decoding, and online serving. To address this issue, we propose UniVA, a Unified Value Alignment framework for advertising recommendation. We first introduce a Commercial SID tokenizer that injects value-related attributes into SID construction, yielding value-discriminative item representations. We then develop a Generation-as-Ranking SID Decoder jointly optimized by supervised learning and eCPM-aware reinforcement learning, which fuses value scores into next-item SID generation to perform generation and ranking in one decoding process. Finally, we design a value-guided personalized beam search that reuses generation-as-ranking logits as online value guidance and applies a personalized trie tree to constrain decoding to request-valid SID paths. Experiments on the Tencent WeChat Channels advertising platform show that UniVA achieves a 37.04\% improvement in offline Hit Rate@100 over the baseline and a 1.5\% GMV lift in online A/B tests.