# RASTP: Representation-Aware Semantic Token Pruning for Generative Recommendation with Semantic Identifiers

arXiv: https://arxiv.org/abs/2511.16943 · HTML: https://mlsanity.com/paper/2511.16943/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Visual token pruning in multimodal models, one of 12 directions in Efficient LLM inference and serving.
- Builds on 9 of its 13 references that are in the map.
- First posted on arXiv 21 Nov 2025.

## Placed in

- Efficient LLM inference and serving / [Visual token pruning in multimodal models](https://mlsanity.com/inference/visual-token-pruning-in-multimodal-models/)

## Builds on (most cited first, 9 of 9)

- [Autoregressive Image Generation using Residual Quantization](https://mlsanity.com/paper/2203.01941/) (2022)
- [Attention Score is not All You Need for Token Importance Indicator in KV Cache Reduction: Value Also Matters](https://mlsanity.com/paper/2406.12335/) (2024)
- [IDGenRec: LLM-RecSys Alignment with Textual ID Learning](https://mlsanity.com/paper/2403.19021/) (2024)
- [Sparse Meets Dense: Unified Generative Recommendations with Cascaded Sparse-Dense Representations](https://mlsanity.com/paper/2503.02453/) (2025)
- [FORGE: Forming Semantic Identifiers for Generative Retrieval in Industrial Datasets](https://mlsanity.com/paper/2509.20904/) (2025)
- [Collaboration of Large Language Models and Small Recommendation Models for Device-Cloud Recommendation](https://mlsanity.com/paper/2501.05647/) (2025)
- [ThinkRec: Thinking-based recommendation via LLM](https://mlsanity.com/paper/2505.15091/) (2025)
- [EARN: Efficient Inference Acceleration for LLM-based Generative Recommendation by Register Tokens](https://mlsanity.com/paper/2507.00715/) (2025)
- [FLAME: A Serving System Optimized for Large-Scale Generative Recommendation with Efficiency](https://mlsanity.com/paper/2509.22681/) (2025)

## Abstract (arXiv)

Generative recommendation systems typically leverage Semantic Identifiers (SIDs), which represent each item as a sequence of tokens that encode semantic information. However, representing item ID with multiple SIDs significantly increases input sequence length, which is a major determinant of computational complexity and memory consumption. While existing efforts primarily focus on optimizing attention computation and KV cache, we propose RASTP (Representation-Aware Semantic Token Pruning), which directly prunes less informative tokens in the input sequence. Specifically, RASTP evaluates token importance by combining semantic saliency, measured via representation magnitude, and attention centrality, derived from cumulative attention weights. Since RASTP dynamically prunes low-information or irrelevant semantic tokens, experiments on three real-world Amazon datasets show that RASTP reduces training time by 26.7%, while maintaining or slightly improving recommendation performance. The code has been open-sourced at https://github.com/Yuzt-zju/RASTP.