RecIS: Sparse to Dense, A Unified Training Framework for Recommendation Models
Not on the map. Why?
Abstract (arXiv)
In this paper, we propose RecIS, a unified Sparse-Dense training framework designed to achieve two primary goals: 1. Unified Framework To create a Unified sparse-dense training framework based on the PyTorch ecosystem that meets the training needs of industrial-grade recommendation models that integrated with large models. 2.System Optimization To optimize the sparse component, offering superior efficiency over the TensorFlow-based recommendation models. The dense component, meanwhile, leverages existing optimization technologies within the PyTorch ecosystem. Currently, RecIS is being used in Alibaba for numerous large-model enhanced recommendation training tasks, and some traditional sparse models have also begun training in it.
Key facts. Cited by 2 papers in the map (data through 14 Oct 2026). Builds on 22 of its 39 references that are in the map. First posted on arXiv 25 Sep 2025.
Builds on
22 of 39 references.
References · 22
- 2023FlashAttention-2: Faster Attention with Better Parallelism and Work Partitioningcited by 2,598
- 2019ZeRO: Memory Optimizations Toward Training Trillion Parameter Modelscited by 1,301
- 2017Mixed Precision Trainingcited by 1,156
- 2017Deep Interest Network for Click-Through Rate Predictioncited by 544
- 2020DCN V2: Improved Deep & Cross Network and Practical Lessons for Web-scale Learning to Rank Systemscited by 327
- 2018xDeepFM: Combining Explicit and Implicit Feature Interactions for Recommender Systemscited by 278
- 2024Actions Speak Louder than Words: Trillion-Parameter Sequential Transducers for Generative Recommendationscited by 188
- 2020Search-based User Interest Modeling with Lifelong Sequential Behavior Data for Click-Through Rate Predictioncited by 177
- 2025OneRec: Unifying Retrieve and Rank with Generative Recommender and Iterative Preference Alignmentcited by 173
- 2021One Model to Serve All: Star Topology Adaptive Recommender for Multi-Domain CTR Predictioncited by 78
- 2024Wukong: Towards a Scaling Law for Large-Scale Recommendationcited by 74
- 2024TWIN V2: Scaling Ultra-Long User Behavior Sequence Modeling for Enhanced CTR Prediction at Kuaishoucited by 30
10 more
- 2023On the Embedding Collapse when Scaling up Recommendation Modelscited by 29
- 2021Scaling Law for Recommendation Models: Towards General-purpose User Representationscited by 26
- 2024Enhancing Taobao Display Advertising with Multimodal Representations: Challenges, Approaches and Insightscited by 23
- 2022Joint Optimization of Ranking and Calibration with Contextualized Hybrid Modelcited by 22
- 2023Scaling User Modeling: Large-scale Online User Representations for Ads Personalization in Metacited by 21
- 2022Merlin HugeCTR: GPU-accelerated Recommender System Training and Inferencecited by 20
- 2022PICASSO: Unleashing the Potential of GPU-centric Training for Wide-and-deep Recommender Systemscited by 17
- 2024HoME: Hierarchy of Multi-Gate Experts for Multi-Task Learning at Kuaishoucited by 16
- 2025Unlocking Scaling Law in Industrial Recommendation Systems with a Three-step Paradigm based Large User Modelcited by 15
- 2024Enhancing CTR Prediction through Sequential Recommendation Pre-training: Introducing the SRP4CTR Frameworkcited by 7
Built on by
Citing papers per year: 2025: 1; 2026: 1.
2 papers.