# Bringing Reasoning to Generative Recommendation Through the Lens of Cascaded Ranking

arXiv: https://arxiv.org/abs/2602.03692 · HTML: https://mlsanity.com/paper/2602.03692/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 3 Feb 2026.

## Built on by (most cited first, 3 of 3)

- [AlpsBench: An LLM Personalization Benchmark for Real-Dialogue Memorization and Preference Alignment](https://mlsanity.com/paper/2603.26680/) (2026)
- [Verifiable Reasoning for LLM-based Generative Recommendation](https://mlsanity.com/paper/2603.07725/) (2026)
- [Echoes in Filter Bubble: Diagnosing and Curing Popularity Bias in Generative Recommenders](https://mlsanity.com/paper/2605.16825/) (2026)

## Abstract (arXiv)

Generative Recommendation (GR) has become a promising end-to-end approach with high FLOPS utilization for resource-efficient recommendation. Despite the effectiveness, we show that current GR models suffer from a critical bias amplification issue, where token-level bias escalates as token generation progresses, ultimately limiting the recommendation diversity and hurting the user experience. By comparing against the key factor behind the success of traditional multi-stage pipelines, we reveal two limitations in GR that can amplify the bias: homogeneous reliance on the encoded history, and fixed computational budgets that prevent deeper user preference understanding. To combat the bias amplification issue, it is crucial for GR to 1) incorporate more heterogeneous information, and 2) allocate greater computational resources at each token generation step. To this end, we propose CARE, a simple yet effective cascaded reasoning framework for debiased GR. To incorporate heterogeneous information, we introduce a progressive history encoding mechanism, which progressively incorporates increasingly fine-grained history information as the generation process advances. To allocate more computations, we propose a query-anchored reasoning mechanism, which seeks to perform a deeper understanding of historical information through parallel reasoning steps. We instantiate CARE on three GR backbones. Empirical results on four datasets show the superiority of CARE in recommendation accuracy, diversity, efficiency, and promising scalability. The codes and datasets are available at https://github.com/Linxyhaha/CARE.