# OxygenREC: An Instruction-Following Generative Framework for E-commerce Recommendation

arXiv: https://arxiv.org/abs/2512.22386 · HTML: https://mlsanity.com/paper/2512.22386/

## Key facts

- Cited by 7 papers in the map (data through 14 Oct 2026).
- Placed in LLM reasoning for recommendation, one of 17 directions in LLM reasoning.
- First posted on arXiv 26 Dec 2025.

## Placed in

- LLM reasoning / [LLM reasoning for recommendation](https://mlsanity.com/lr/llm-reasoning-for-recommendation/)

## Built on by (most cited first, 7 of 7)

- [OneReason Technical Report](https://mlsanity.com/paper/2606.06260/) (2026)
- [TwiSTAR:Think Fast, Think Slow, Then Act,Generative Recommendation with Adaptive Reasoning](https://mlsanity.com/paper/2605.11553/) (2026)
- [EvoRec: Self Evolving Agentic Recommender Systems](https://mlsanity.com/paper/2606.28368/) (2026)
- [Think2Go: Generative Next POI Recommendation with LLM Reasoning](https://mlsanity.com/paper/2607.28997/) (2026)
- [OCP: Orthogonal Constrained Projection for Sparse Scaling in Industrial Commodity Recommendation](https://mlsanity.com/paper/2603.18697/) (2026)
- [AIGQ: An End-to-End Hybrid Generative Architecture for E-commerce Query Recommendation](https://mlsanity.com/paper/2603.19710/) (2026)
- [ReCast: Recasting Learning Signals for Reinforcement Learning in Generative Recommendation](https://mlsanity.com/paper/2604.22169/) (2026)

## Abstract (arXiv)

Traditional recommendation systems suffer from inconsistency in multi-stage optimization objectives. Generative Recommendation (GR) mitigates them through an end-to-end framework; however, existing methods still rely on matching mechanisms based on inductive patterns. Although responsive, they lack the ability to uncover complex user intents that require deductive reasoning based on world knowledge. Meanwhile, LLMs show strong deep reasoning capabilities, but their latency and computational costs remain challenging for industrial applications. More critically, there are performance bottlenecks in multi-scenario scalability: as shown in Figure 1, existing solutions require independent training and deployment for each scenario, leading to low resource utilization and high maintenance costs-a challenge unaddressed in GR literature. To address these, we present OxygenREC, an industrial recommendation system that leverages Fast-Slow Thinking to deliver deep reasoning with strict latency and multi-scenario requirements of real-world environments. First, we adopt a Fast-Slow Thinking architecture. Slow thinking uses a near-line LLM pipeline to synthesize Contextual Reasoning Instructions, while fast thinking employs a high-efficiency encoder-decoder backbone for real-time generation. Second, to ensure reasoning instructions effectively enhance recommendation generation, we introduce a semantic alignment mechanism with Instruction-Guided Retrieval (IGR) to filter intent-relevant historical behaviors and use a Query-to-Item (Q2I) loss for instruction-item consistency. Finally, to resolve multi-scenario scalability, we transform scenario information into controllable instructions, using unified reward mapping and Soft Adaptive Group Clip Policy Optimization (SA-GCPO) to align policies with diverse business objectives, realizing a train-once-deploy-everywhere paradigm.