Generative Recommendation (GR) reframes retrieval and ranking as autoregressive decoding over Semantic IDs (SIDs), unifying the multi-stage pipeline into a single model. Yet a fundamental expressive gap persists: discriminative models score items with direct feature access, enabling explicit user-item crossing, whereas GR decodes over compact SID tokens without item-side signal. We formalize this via Bayes' theorem, showing ranking by p(y|f,u) is equivalent to ranking by p(f|y,u), which factorizes autoregressively over item features. This establishes that a generative model with full feature access is as expressive as its discriminative counterpart; any practical gap stems solely from incomplete feature coverage. We propose UniRec with Chain-of-Attribute (CoA) as its core mechanism. CoA prefixes each SID sequence with structured attribute tokens--category, seller, brand--before decoding the SID itself, recovering the item-side feature crossing that discriminative models exploit. Because items sharing identical attributes cluster in adjacent SID regions, attribute conditioning yields a measurable per-step entropy reduction H(s_k|s_{<k},a) < H(s_k|s_{<k}), narrowing the search space and stabilizing beam search trajectories. We further address two deployment challenges: Capacity-constrained SID introduces exposure-weighted capacity penalties into residual quantization to suppress token collapse and the Matthew effect across SID layers; Conditional Decoding Context (CDC) combines Task-Conditioned BOS with hash-based Content Summaries, injecting scenario-conditioned signals at each decoding step. A joint RFT and DPO framework aligns the model with business objectives beyond distribution matching. Experiments show UniRec outperforms the strongest baseline by +22.6% HR@50 overall and +15.5% on high-value orders, with online A/B tests confirming significant business metric gains.

翻译：生成式推荐（GR）将检索与排序重构为基于语义ID（SID）的自回归解码，将多阶段流水线统一为单一模型。然而，根本性的表达性鸿沟依然存在：判别式模型通过直接特征访问对物品进行评分，支持显式的用户-物品交叉；而GR仅基于紧凑的SID令牌进行解码，缺乏物品端信号。我们通过贝叶斯定理对此进行形式化分析，证明基于p(y|f,u)排序等价于基于p(f|y,u)排序，后者可对物品特征进行自回归分解。这表明具备完全特征访问能力的生成模型与判别式模型具有同等表达性；任何实际鸿沟仅源于不完整的特征覆盖。我们提出UniRec，其核心机制为属性链（CoA）。CoA在解码SID序列前，为每个序列添加结构化属性令牌——类别、卖家、品牌——从而恢复判别式模型所利用的物品端特征交叉。由于共享相同属性的物品在SID序列中相邻聚类，属性条件化可实现可量化的逐步骤熵减H(s_k|s_{<k},a) < H(s_k|s_{<k})，从而缩小搜索空间并稳定束搜索轨迹。我们进一步解决两个部署挑战：容量受限的SID通过向残差量化引入曝光加权容量惩罚，以抑制SID层间的令牌坍塌和马太效应；条件解码上下文（CDC）结合任务条件化起始符与基于哈希的内容摘要，在每一步解码中注入场景条件化信号。联合RFT与DPO框架使模型对齐商业目标，超越分布匹配。实验表明UniRec在HR@50指标上超越最强基线+22.6%，在高价值订单上提升+15.5%，在线A/B测试验证了显著的业务指标收益。