Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

WarpRec: Unifying Academic Rigor and Industrial Scale for Responsible, Reproducible, and Efficient Recommendation

Marco Avolio, Potito Aghilar, Sabino Roccotelli, Vito Walter Anelli, Chiara Mallamaci, Vincenzo Paparella, Marco Valentini, Alejandro Bellogín, Michelantonio Trizio, Joseph Trotta, Antonio Ferrara, Tommaso Di Noia

TL;DR

WarpRec addresses the deployment chasm in recommender systems by delivering a backend-agnostic, modular framework that scales from local prototyping to distributed training while embedding CodeCarbon-based Green AI measures. Its five-engine architecture (Reader, Data Engine, Recommendation Engine, Writer, Application Layer) paired with Narwhals enables write-once, run-anywhere experimentation and rigorous evaluation across 40 metrics with statistical corrections. The framework supports 55 models across diverse paradigms, 13 filtering strategies, and 6 data-splitting strategies, all under declarative pipelines and agent-ready MCP interfaces, bridging scientific rigor, industrial scale, and agentic AI readiness. Empirical benchmarks demonstrate WarpRec’s robustness across scales, superior end-to-end performance, and notable energy and carbon profiles, underscoring its practical impact for sustainable, interactive RS in real-world settings.

Abstract

Innovation in Recommender Systems is currently impeded by a fractured ecosystem, where researchers must choose between the ease of in-memory experimentation and the costly, complex rewriting required for distributed industrial engines. To bridge this gap, we present WarpRec, a high-performance framework that eliminates this trade-off through a novel, backend-agnostic architecture. It includes 50+ state-of-the-art algorithms, 40 metrics, and 19 filtering and splitting strategies that seamlessly transition from local execution to distributed training and optimization. The framework enforces ecological responsibility by integrating CodeCarbon for real-time energy tracking, showing that scalability need not come at the cost of scientific integrity or sustainability. Furthermore, WarpRec anticipates the shift toward Agentic AI, leading Recommender Systems to evolve from static ranking engines into interactive tools within the Generative AI ecosystem. In summary, WarpRec not only bridges the gap between academia and industry but also can serve as the architectural backbone for the next generation of sustainable, agent-ready Recommender Systems. Code is available at https://github.com/sisinflab/warprec/

WarpRec: Unifying Academic Rigor and Industrial Scale for Responsible, Reproducible, and Efficient Recommendation

TL;DR

WarpRec addresses the deployment chasm in recommender systems by delivering a backend-agnostic, modular framework that scales from local prototyping to distributed training while embedding CodeCarbon-based Green AI measures. Its five-engine architecture (Reader, Data Engine, Recommendation Engine, Writer, Application Layer) paired with Narwhals enables write-once, run-anywhere experimentation and rigorous evaluation across 40 metrics with statistical corrections. The framework supports 55 models across diverse paradigms, 13 filtering strategies, and 6 data-splitting strategies, all under declarative pipelines and agent-ready MCP interfaces, bridging scientific rigor, industrial scale, and agentic AI readiness. Empirical benchmarks demonstrate WarpRec’s robustness across scales, superior end-to-end performance, and notable energy and carbon profiles, underscoring its practical impact for sustainable, interactive RS in real-world settings.

Abstract

Innovation in Recommender Systems is currently impeded by a fractured ecosystem, where researchers must choose between the ease of in-memory experimentation and the costly, complex rewriting required for distributed industrial engines. To bridge this gap, we present WarpRec, a high-performance framework that eliminates this trade-off through a novel, backend-agnostic architecture. It includes 50+ state-of-the-art algorithms, 40 metrics, and 19 filtering and splitting strategies that seamlessly transition from local execution to distributed training and optimization. The framework enforces ecological responsibility by integrating CodeCarbon for real-time energy tracking, showing that scalability need not come at the cost of scientific integrity or sustainability. Furthermore, WarpRec anticipates the shift toward Agentic AI, leading Recommender Systems to evolve from static ranking engines into interactive tools within the Generative AI ecosystem. In summary, WarpRec not only bridges the gap between academia and industry but also can serve as the architectural backbone for the next generation of sustainable, agent-ready Recommender Systems. Code is available at https://github.com/sisinflab/warprec/
Paper Structure (16 sections, 2 figures, 5 tables)

This paper contains 16 sections, 2 figures, 5 tables.

Table of Contents

  1. Introduction
  2. The Fragmented Landscape
  3. WarpRec
  4. Pipelines and Callbacks
  5. Reader Module
  6. Data Engine
  7. Recommendation Engine
  8. Evaluation
  9. Writer Module
  10. Application Layer
  11. Performance and Multi-faceted Analysis
  12. Experimental Setup
  13. RQ1: Scalability and Performance
  14. RQ2: Green AI Analysis
  15. RQ3: Interoperability with Agentic Pipelines
  16. ...and 1 more sections

Figures (2)

  • Figure 1: The modular architecture of WarpRec. Five decoupled modules manage the end-to-end recommendation lifecycle, from data ingestion and processing to model training and evaluation. An Application Layer exposes the recommender through a REST API and MCP agentic interface.
  • Figure 2: Sequential Recommendation via WarpRec MCP Interface. The AI Agent leverages a SASRec model trained on MovieLens-32M.