ACL Findings2025
AGRec: Adapting Autoregressive Decoders with Graph Reasoning for LLM-based Sequential Recommendation
Xinfeng Wang, Jin Cui, Fumiyo Fukumoto, Yoshimi Suzuki
Abstract
Autoregressive decoders in large language models (LLMs) excel at capturing users' sequential behaviors for generative recommendations. However, they inherently struggle to leverage graph-structured user-item interactions, which are widely recognized as beneficial. This paper presents AGRec, adapting LLMs' decoders with graph reasoning for recommendation. We reveal that LLMs and graph neural networks (GNNs) manifest complementary strengths in distinct user domains. Building on this, we augment the decoding logits of LLMs with an auxiliary GNN model to optimize token generation. Moreover, we introduce a rankable finite state machine to tackle two challenges: (1) adjusting autoregressive generation with discriminative decoders that directly predict user-item similarity, and (2) token homogeneity, where LLMs often generate items with similar prefix tokens, narrowing the scope of beam search. Our AGRec outperforms state-of-the-art models in sequential recommendations. Our source code and data are available online 1 .
Ask Lune
Know this paper in minutes, not an afternoon.
Lune can read every page of it and the top-tier papers that cite it, and gives you its claims, its evidence and its limits, each one cited to the paper it comes from. No skimming, no AI slop.
Lineage
- Self-supervised Graph Learning for Recommendation cites LightGCN: Simplifying and Powering Graph Convolution Network for Recommendation
- Are Graph Augmentations Necessary?: Simple Graph Contrastive Learning for Recommendation cites LightGCN: Simplifying and Powering Graph Convolution Network for Recommendation
- Are Graph Augmentations Necessary?: Simple Graph Contrastive Learning for Recommendation cites Self-supervised Graph Learning for Recommendation
Builds on28
- LoRA: Low-Rank Adaptation of Large Language ModelsEdward J. Hu, Yelong Shen, Phillip Wallis, Zeyuan Allen-Zhu et al.ICLR 2022 · 18,833 citations
- LightGCN: Simplifying and Powering Graph Convolution Network for RecommendationXiangnan He, Kuan Deng, Xiang Wang, Yan Li et al.SIGIR 2020 · 4,448 citations
- Self-supervised Graph Learning for RecommendationJiancan Wu, Xiang Wang, Fuli Feng, Xiangnan He et al.SIGIR 2021 · 1,476 citations
- Are Graph Augmentations Necessary?: Simple Graph Contrastive Learning for RecommendationJunliang Yu, Hongzhi Yin, Xin Xia, Tong Chen et al.SIGIR 2022 · 658 citations
- Recommender Systems with Generative RetrievalShashank Rajput, Nikhil Mehta, Anima Singh, Raghunandan Hulikal Keshavan et al.NeurIPS 2023 · 474 citations
Related papers
- Enhancing High-order Interaction Awareness in LLM-based Recommender ModelXinfeng Wang, Jin Cui, Fumiyo Fukumoto, Yoshimi SuzukiEMNLP 2024 · 6 citations
- UCGRec: User-Centric Graph Learning for LLM-based Sequential RecommendationHanBeul Kim, CheolWon Na, Suyoung Bae, YunSeok Choi et al.ACL Findings 2026
- LLMRG: Improving Recommendations through Large Language Model Reasoning GraphsYan Wang, Zhixuan Chu, Xin Ouyang, Simeng Wang et al.AAAI 2024 · 47 citations
- CORONA: A Coarse-to-Fine Framework for Graph-based Recommendation with Large Language ModelsJunze Chen, Xinjie Yang, Cheng Yang, Junfei Bao et al.SIGIR 2025 · 5 citations
- Enhancing Large Language Model Based Sequential Recommender Systems with Pseudo Labels ReconstructionHyunsoo Na, Minseok Gang, Youngrok Ko, Jinseok Seol et al.EMNLP Findings 2024 · 3 citations
