---
title: "HiGenTrans: Hierarchical LLM Knowledge Transfer for Generative Recommendation"
source: "arXiv AI/CL/LG（论文）"
source_tier: "t2"
url: "https://arxiv.org/abs/2611.01129v1"
permalink: "https://feed.astralor.com/items/itm_9691cfd61ff4eb77222ef5c1"
published: "2026-10-31T17:19:23.000Z"
score: 58
channel: "tech"
tags: ["LLM","多模态","HiGenTrans","Hierarchical LLM Knowledge","Transfer","Generative Recommendation Generative","Semantic IDs","SIDs"]
---

# HiGenTrans: Hierarchical LLM Knowledge Transfer for Generative Recommendation

## 速览

Generative recommendation promises to simplify conventional multi-stage recommender systems by generating item Semantic IDs (SIDs) autoregressively. However, two obstacles limit its effective- ness in industrial setting…

## 关键点

- Generative recommendation promises to simplify conventional multi-stage recommender systems by generating item Semantic IDs (SIDs) autoregressively.
- However, two obstacles limit its effective- ness in industrial settings: interaction-only supervision underuses the semantic knowledge encoded in pretrained language models, while beam-search likelihood ranks token sequ…
- To address these limitations, we propose HiGenTrans, a three-level framework that introduces LLM knowledge at the input, representation, and decision levels.

## 全文

Generative recommendation promises to simplify conventional multi-stage recommender systems by generating item Semantic IDs (SIDs) autoregressively. However, two obstacles limit its effective- ness in industrial settings: interaction-only supervision underuses the semantic knowledge encoded in pretrained language models, while beam-search likelihood ranks token sequences rather than directly estimating item relevance. To address these limitations, we propose HiGenTrans, a three-level framework that introduces LLM knowledge at the input, representation, and decision levels. First, Global Semantic Priming (GSP) adds an LLM-derived user representation to the beginning-of-sequence embedding, provid- ing a semantic prior before SID generation. Second, Hierarchi- cal Representation Transfer (HRT) aligns intermediate LLM representations with depth-matched GenRec layers through mean- centered projections and gated residual connections. Finally, Multi- Signal Decision Calibration (MDC) uses a lightweight mixture- of-experts calibrator to combine semantic relevance, generation confidence, and candidate-set context when reranking beam-search outputs. All semantic features are computed only from the observ- able interaction prefix, preventing target leakage, and can be cached offline. Experiments on three public benchmarks show that HiGen- Trans improves Recall@10 by up to 12.9% and NDCG@10 by up to 33.7% over the strongest baselines. These results demonstrate the value of transferring LLM knowledge at complementary stages of the generative recommendation pipeline.

---

引用前请核对原文;本文由星际流动整理:https://feed.astralor.com/items/itm_9691cfd61ff4eb77222ef5c1