Aug 2026· ACM Transactions on Information Systems· 0 citations· 58 references
TL;DR
This work proposes RosePO, a framework to refine LLM-based recommendation through pairwise preference optimization with personalized smoothing, and incorporates a personalized smoothing factor predicted by a user oracle into the optimization objective.
Abstract
Recent advancements in Large Language Models (LLMs) have inspired their application in sequential recommendation systems, often through supervised fine-tuning (SFT). However, conventional SFT methods often struggle to capture nuanced comparative relationships between items. While recent approaches utilize Direct Preference Optimization (DPO), they remain constrained by challenges in modeling customized preferences, including capturing fine-grained user preferences and being susceptible to semantic hallucination and popularity bias. To overcome these challenges, we propose RosePO, a framework to refine LLM-based recommendation through pairwise preference optimization with personalized smoothing. We illustrate the concept with three critical preference examples pertinent to LLM-based recommendation. Specifically, we design rejected sampling strategies tailored for each customized preference. To ensure robustness against uncertain labels present in automatically constructed preference data, we incorporate a personalized smoothing factor predicted by a user oracle into the optimization objective. Empirical evaluation on three real-world datasets demonstrates the effectiveness of our method, showcasing not only promising recommendation performance but also mitigation of semantic hallucination and popularity bias. We hope this work paves a way to build helpful and harmless LLM-based recommendation service in the future.
This work proposes a novel Token Selection approach for Preference Optimization in LLM-based sequential Recommendation, i.e., TSPORec, which accurately pinpoints informative tokens throughout the entire textual content to improve recommendation performance.
Wenqiao Zhu, Chao Xu, Haipang Wu et al.· 0 citations
Large Language Models (LLMs) have demonstrated remarkable general-purpose abilities across a wide range of domains, and these strengths have also been increasingly evidenced in recommender systems. However, existing methods that attempt to integrate collaborative signals into LLMs often fail to preserve their foundational knowledge. This loss is critical in text-rich recommendation, where robust semantic understanding is required to interpret user reviews and item profiles. We propose PALRec, a parameter-preserving augmentation framework that equips an LLM with recommendation capabilities while keeping its original parameters fixed. We first construct evidence-grounded user and item profiles from reviews and use them as concise pseudo-labels for reconstruction. We then introduce lightweight, trainable user and item embedding modules optimized with a multi-task objective that combines next-item prediction and profile reconstruction. These modules are trained jointly to align collaborative signals with the LLM’s semantic space without modifying the backbone. We also employ token-aware loss decomposition and frequency-aware reweighting to stabilize training and mitigate popularity bias. Experiments on public benchmarks show that PALRec consistently outperforms fully fine-tuned counterparts in recommendation accuracy while preserving the LLM’s pre-trained knowledge. This result highlights that maintaining the LLM’s semantic understanding is crucial for effectively exploiting textual information in recommender systems.
Hyunsoo Na, Minseok Gang, Sang-goo Lee et al.· ACM Transactions on Informat...· 0 citations
Large language models (LLM) have emerged as a pivotal technique for enhancing the performance of modern recommender systems. Despite enjoying many advantages, LLM-based recommender systems (LRS) exhibit more severe popularity bias than conventional recommender systems (CRS). Compounding this issue, current debiasing methods for LRS are limited to single-level and single-stage interventions, resulting in suboptimal effectiveness. In this paper, we propose a novel LRS debiasing method that performs debiasing at both the token and item levels across the training and inference stages, so called Hierarchical Dual Debiasing in LRS (HD \({}^{2}\) LRS). Specifically, during the training stage, a token-wise inverse propensity score technique is embedded into the supervised fine-tuning phase, encouraging the LLM to focus more on unpopular tokens. Subsequently, a popularity-aware direct preference optimization method is developed to curb LLM's tendency to favor popular items. During the inference stage, we first train a highly biased CRS by amplifying the weights assigned to popular items. The prediction outputs of the biased CRS are then leveraged to constrain the LLM, preventing it from generating popular tokens and recommending popular items. Extensive experiments conducted on four real-world datasets confirm that our HD \({}^{2}\) LRS effectively alleviates popularity bias in LRS and yields significant improvements in recommendation accuracy.
Sijin Lu, Jun Wu· ACM Transactions on Informat...· 0 citations
The Fusion of Layer-wise Exits for Sequential Recommendation (FLEXRec), a discriminative framework that enhances compact LLMs while retaining scalable full-corpus ranking and achieves state-of-the-art accuracy among competing methods while remaining highly efficient.
Xurong Liang, Tong Chen, Q. Nguyen et al.· 0 citations
Large language models (LLMs) have shown promise for training-free recommendation, but LLM-generated user interests are often too broad for fine-grained item retrieval. Existing methods incorporate collaborative filtering (CF) signals in a pre-LLM manner through candidate reranking or prompt augmentation, yielding limited gains. We propose CoRRe, a training-free recommendation framework with a post-LLM paradigm that injects CF signals into LLM-generated item representations, which are later matched with LLM-generated user interests for ranking. Specifically, CoRRe refines the directions of item embeddings using an item-item co-purchase graph and their magnitudes using item popularity. Experiments on real-world datasets show that CoRRe consistently outperforms existing training-free methods and achieves competitive or superior performance compared with training-based methods, without requiring any model training or task-specific fine-tuning.
Kyungho Kim, Sunwoo Kim, Geon Lee et al.· 0 citations
