OneRank: Unified Transformer-Native Ranking Architecture for Multi-Task Recommendation
2026-06-15 • Information Retrieval
Information Retrieval
AI summaryⓘ
The authors point out that current recommendation systems use Transformers mainly as general feature encoders, which limits how well they can learn multiple user tasks at once. They propose OneRank, a new method that mixes feature encoding and task prediction inside the Transformer itself, allowing each task to have its own dedicated learning paths while still sharing useful information. This design helps avoid problems like conflicting updates between tasks and makes the model better at personalizing recommendations. Tests on real-world data showed OneRank works better than existing methods and is efficient to run.
Multi-task learningTransformerRecommender systemsGradient interferenceTask-private channelsRepresentation learningPersonalized rankingGradient detachmentMatching-based scoringContextualization
Authors
Jiakai Tang, Sunhao Dai, Kun Wang, Zhiluohan Guo, Yu Zhao, Cong Fu, Kangle Wu, Yabo Ni, Anxiang Zeng, Xu Chen, Jun Xu
Abstract
Multi-task learning (MTL) is essential in recommender systems to enable complementary learning among diverse user feedback. While modern industrial practices have shifted from DNNs to Transformer-centric architectures to strengthen sequence modeling and scaling capacity, they still decouple feature encoding from multi-task prediction, treating the Transformer as a task-agnostic encoder. This design fundamentally limits the performance and scalability by (1) creating an information bottleneck under heterogeneous task objectives, (2) inducing gradient interference that leads to the seesaw phenomenon, and (3) forcing a dataflow transition in which attention-based, context-adaptive representation learning is converted to static feed-forward task prediction with incompatible information read-write dynamics. We propose OneRank, a Transformer-native multi-task ranking framework that eliminates encoder-predictor separation and introduces task-private channels for forward representation learning and backward optimization, enabling task-specialized learning while reducing inter-task interference. In the forward pass, OneRank learns task-specific representations bottom-up through task-conditioned information selection, candidate-aware contextualization, and controlled cross-task interaction. In the backward pass, cross-task gradient detachment isolates task-private parameter updates from shared knowledge extraction modules, preventing negative transfer. We further replace static task-specific MLP scorers with dynamic matching-based scoring for context-aware personalized ranking. By internalizing multi-task reasoning within the Transformer stack, OneRank establishes a unified and scalable architectural paradigm. Offline and online experiments on large-scale industrial datasets show that OneRank significantly outperforms state-of-the-art baselines while maintaining computational efficiency.