# FineSID: Scalable and Efficient Semantic Identifier Learning for Generative Recommendation

arXiv: https://arxiv.org/abs/2609.36670 · HTML: https://mlsanity.com/paper/2609.36670/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 23 of its 32 references that are in the map.
- First posted on arXiv 29 Sep 2026.

## Builds on (most cited first, 23 of 23)

- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [LLaMA: Open and Efficient Foundation Language Models](https://mlsanity.com/paper/2302.13971/) (2023)
- [Self-Attentive Sequential Recommendation](https://mlsanity.com/paper/1808.09781/) (2018)
- [BERT4Rec: Sequential Recommendation with Bidirectional Encoder Representations from Transformer](https://mlsanity.com/paper/1904.06690/) (2019)
- [Autoregressive Image Generation using Residual Quantization](https://mlsanity.com/paper/2203.01941/) (2022)
- [Session-based Recommendations with Recurrent Neural Networks](https://mlsanity.com/paper/1511.06939/) (2015)
- [Personalized Top-N Sequential Recommendation via Convolutional Sequence Embedding](https://mlsanity.com/paper/1809.07426/) (2018)
- [Adapting Large Language Models by Integrating Collaborative Semantics for Recommendation](https://mlsanity.com/paper/2311.09049/) (2023)
- [RecBole: Towards a Unified, Comprehensive and Efficient Framework for Recommendation Algorithms](https://mlsanity.com/paper/2011.01731/) (2020)
- [Learnable Item Tokenization for Generative Recommendation](https://mlsanity.com/paper/2405.07314/) (2024)
- [Hierarchical Gating Networks for Sequential Recommendation](https://mlsanity.com/paper/1906.09217/) (2019)
- [From Matching to Generation: A Survey on Generative Information Retrieval](https://mlsanity.com/paper/2404.14851/) (2024)
- [Learning to Tokenize for Generative Retrieval](https://mlsanity.com/paper/2304.04171/) (2023)
- [RecBole 2.0: Towards a More Up-to-Date Recommendation Library](https://mlsanity.com/paper/2206.07351/) (2022)
- [Norm of Word Embedding Encodes Information Gain](https://mlsanity.com/paper/2212.09663/) (2022)
- [MiniOneRec: An Open-Source Framework for Scaling Generative Recommendation](https://mlsanity.com/paper/2510.24431/) (2025)
- [Generative Retrieval with Preference Optimization for E-commerce Search](https://mlsanity.com/paper/2407.19829/) (2024)
- [Unified Semantic and ID Representation Learning for Deep Recommenders](https://mlsanity.com/paper/2502.16474/) (2025)
- [Understanding Generative Recommendation with Semantic IDs from a Model-scaling View](https://mlsanity.com/paper/2509.25522/) (2025)
- [Contrastive Learning-based Sentence Encoders Implicitly Weight Informative Words](https://mlsanity.com/paper/2310.15921/) (2023)
- [Act-With-Think: Chunk Auto-Regressive Modeling for Generative Recommendation](https://mlsanity.com/paper/2506.23643/) (2025)
- [Efficient Recommendation with Millions of Items by Dynamic Pruning of Sub-Item Embeddings](https://mlsanity.com/paper/2505.00560/) (2025)
- [PROMISE: Process Reward Models Unlock Test-Time Scaling Laws in Generative Recommendations](https://mlsanity.com/paper/2601.04674/) (2026)

## Abstract (arXiv)

A critical prerequisite of generative recommendation is designing semantic identifiers (SIDs) that are both scalable to large item sets and efficiently learnable. Existing SID learning methods fundamentally rely on Top-1 hard assignment during vector quantization. While heuristic strategies -- such as clustering-based initialization or forced post-hoc collision resolution -- can artificially inflate codebook coverage, they often disrupt end-to-end semantic alignment and fail to address the underlying optimization bottleneck: sparse gradient propagation. In standard Top-1 assignment, gradients concentrate on a narrow subset of frequently selected codewords, leaving the majority inherently under-trained and causing severe SID collisions. To overcome this limitation natively without relying on complex initialization priors, we propose FineSID, a unified quantization framework that moves beyond Top-1 assignment by enabling fine-grained gradient propagation across the entire codebook. Instead of updating only a single selected codeword, FineSID distributes learning signals to all codewords in a soft, differentiable manner. This design promotes globally balanced codebook optimization while strictly preserving semantic consistency, effectively alleviating SID collisions and stabilizing training in large, high-dimensional codebooks. Extensive experiments on multiple public benchmarks demonstrate that FineSID is robust to initialization configurations and consistently improves both codebook utilization and recommendation accuracy. Our work provides a principled, initialization-agnostic solution for semantic identifier learning, advancing the practicality of generative recommendation.