arXiv subjects

Yingjie Qin

Publications and source records attributed to Yingjie Qin.

8 recordsLinked to original sources

arXiv2026

MOTOR: Learning ID-free Item Representation with Token Crossing for Embedding-based Multimodal Recommendation

While multimodal recommendation models have effectively integrated visual and textual information, their reliance on unique ID embeddings constitutes a fundamental performance bottleneck. Specifically, ID-based paradigms suffer from three limitations: (1) \textbf{Information Isolation}, where unique IDs prevent semantic information exchange among related items; (2) \textbf{Cold-Start Vulnerability}, as ID embeddings are difficult to optimize with sparse interactions; and (3) \textbf{Storage Inefficiency}, where parameter costs scale linearly with item quantity. To overcome these challenges, we propose \textbf{MOTOR}, a novel \textbf{ID-free MultimOdal TOken Representation} scheme. MOTOR replaces explicit item IDs with learnable, shared multimodal tokens, fundamentally transforming the recommender into an ID-free framework. Methodologically, we first employ product quantization to discretize raw multimodal features into compact token IDs. These tokens serve as implicit item features, which are then synthesized via a novel \textbf{Token Cross Network (TCN)} to capture high-order interaction patterns. This "discretize-and-interact" mechanism enables semantic sharing across items and significantly compresses the model size without introducing complex auxiliary losses. Extensive experiments across nine mainstream models demonstrate the significant performance improvement achieved by MOTOR. Further, MOTOR improves the capability of these models to recommend items in cold-start scenarios.

cs.IR↗

arXiv2026

DiffCold: A Diffusion-based Generative Model for Cold-Start Item Recommendation

Cold-start item recommendation remains a persistent challenge in real-world systems due to the absence of interaction histories. While prior models attempt to bridge this gap using item content features, they universally suffer from the \textbf{seesaw dilemma}: enhancing performance for cold items inevitably degrades performance for warm items, and vice versa. We identify that this dilemma stems from a fundamental \textbf{distributional disparity}: warm item embeddings occupy a complex ``behavioral manifold" shaped by rich interaction signals, whereas cold item embeddings are constrained to a ``semantic manifold" derived solely from auxiliary content. Existing methods often force a rigid mapping between these inconsistent spaces, causing the model to sacrifice the precision of warm representations to accommodate cold ones. To address this, we propose \textbf{DiffCold}, a diffusion-based generative model that unifies warm and cold representations. Unlike GANs or VAEs, DiffCold leverages conditional diffusion to reconstruct warm item embeddings from content, preserving the underlying manifold structure without degradation. We further tailor this paradigm with two specific designs: a \textbf{Retrieval-enhanced Aggregator} that initializes generation using semantically similar warm items to bypass inefficient noise, and a \textbf{Simulation-based Representation Alignment} module that enforces distribution consistency between generated and real embeddings via contrastive learning. Experiments on three benchmarks confirm that DiffCold resolves the seesaw dilemma, consistently outperforming state-of-the-art methods across all metrics.

cs.IR↗

arXiv2026

LASER: An Efficient Target-Aware Segmented Attention Framework for End-to-End Long Sequence Modeling

Modeling ultra-long user behavior sequences is pivotal for capturing evolving and lifelong interests in modern recommendation systems. However, deploying such models in real-time industrial environments faces a strict "Latency Wall", constrained by two distinct bottlenecks: the high I/O latency of retrieving massive user histories and the quadratic computational complexity of standard attention mechanisms. To break these bottlenecks, we present LASER, a full-stack optimization framework developed and deployed at Xiaohongshu (RedNote). Our approach tackles the challenges through two complementary innovations: (1) System efficiency: We introduce SeqVault, a unified schema-aware serving infrastructure for long user histories. By implementing a hybrid DRAM-SSD indexing strategy, SeqVault reduces retrieval latency by 50% and CPU usage by 75%, ensuring millisecond-level access to full real-time and life-cycle user histories. (2) Algorithmic efficiency: We propose a Segmented Target Attention (STA) mechanism to address the computational overhead. Motivated by the inherent sparsity of user interests, STA employs a sigmoid-based gating strategy that acts as a silence mechanism to filter out noisy items. Subsequently, a lightweight Global Stacked Target Attention (GSTA) module refines these compressed segments to capture cross-segment dependencies without incurring high computational costs. This design performs effective sequence compression, reducing the complexity of long-sequence modeling while preserving critical signals. Extensive offline evaluations demonstrate that LASER consistently outperforms state-of-the-art baselines. In large-scale online A/B testing serving over 100 million daily active users, LASER achieved a 2.36% lift in ADVV and a 2.08% lift in revenue, demonstrating its scalability and significant commercial impact.

cs.IR↗

arXiv2025

HoPE: Hybrid of Position Embedding for Long Context Vision-Language Models

Vision-Language Models (VLMs) have made significant progress in multimodal tasks. However, their performance often deteriorates in long-context scenarios, particularly long videos. While Rotary Position Embedding (RoPE) has been widely adopted for length generalization in Large Language Models (LLMs), extending vanilla RoPE to capture the intricate spatial-temporal dependencies in videos remains an unsolved challenge. Existing methods typically allocate different frequencies within RoPE to encode 3D positional information. However, these allocation strategies mainly rely on heuristics, lacking in-depth theoretical analysis. In this paper, we first study how different allocation strategies impact the long-context capabilities of VLMs. Our analysis reveals that current multimodal RoPEs fail to reliably capture semantic similarities over extended contexts. To address this issue, we propose HoPE, a Hybrid of Position Embedding designed to improve the long-context capabilities of VLMs. HoPE introduces a hybrid frequency allocation strategy for reliable semantic modeling over arbitrarily long contexts, and a dynamic temporal scaling mechanism to facilitate robust learning and flexible inference across diverse context lengths. Extensive experiments across four video benchmarks on long video understanding and retrieval tasks demonstrate that HoPE consistently outperforms existing methods, confirming its effectiveness. Our code is available at https://github.com/hrlics/HoPE.

cs.LG↗

arXiv2025

GIST: Cross-Domain Click-Through Rate Prediction via Guided Content-Behavior Distillation

Cross-domain Click-Through Rate prediction aims to tackle the data sparsity and the cold start problems in online advertising systems by transferring knowledge from source domains to a target domain. Most existing methods rely on overlapping users to facilitate this transfer, often focusing on joint training or pre-training with fine-tuning approach to connect the source and target domains. However, in real-world industrial settings, joint training struggles to learn optimal representations with different distributions, and pre-training with fine-tuning is not well-suited for continuously integrating new data. To address these issues, we propose GIST, a cross-domain lifelong sequence model that decouples the training processes of the source and target domains. Unlike previous methods that search lifelong sequences in the source domains using only content or behavior signals or their simple combinations, we innovatively introduce a Content-Behavior Joint Training Module (CBJT), which aligns content-behavior distributions and combines them with guided information to facilitate a more stable representation. Furthermore, we develop an Asymmetric Similarity Integration strategy (ASI) to augment knowledge transfer through similarity computation. Extensive experiments demonstrate the effectiveness of GIST, surpassing SOTA methods on offline evaluations and an online A/B test. Deployed on the Xiaohongshu (RedNote) platform, GIST effectively enhances online ads system performance at scale, serving hundreds of millions of daily active users.

cs.AI↗

arXiv2025

Diffusion Model for Interest Refinement in Multi-Interest Recommendation

Multi-interest candidate matching plays a pivotal role in personalized recommender systems, as it captures diverse user interests from their historical behaviors. Most existing methods utilize attention mechanisms to generate interest representations by aggregating historical item embeddings. However, these methods only capture overall item-level relevance, leading to coarse-grained interest representations that include irrelevant information. To address this issue, we propose the Diffusion Multi-Interest model (DMI), a novel framework for refining user interest representations at the dimension level. Specifically, DMI first introduces controllable noise into coarse-grained interest representations at the dimensional level. Then, in the iterative reconstruction process, DMI combines a cross-attention mechanism and an item pruning strategy to reconstruct the personalized interest vectors with the guidance of tailored collaborative information. Extensive experiments demonstrate the effectiveness of DMI, surpassing state-of-the-art methods on offline evaluations and an online A/B test. Successfully deployed in the real-world recommender system, DMI effectively enhances user satisfaction and system performance at scale, serving the major traffic of hundreds of millions of daily active users. \footnote{The code will be released for reproducibility once the paper is accepted.}

cs.IR↗

arXiv2024

DREAM: A Dual Representation Learning Model for Multimodal Recommendation

Multimodal recommendation focuses primarily on effectively exploiting both behavioral and multimodal information for the recommendation task. However, most existing models suffer from the following issues when fusing information from two different domains: (1) Previous works do not pay attention to the sufficient utilization of modal information by only using direct concatenation, addition, or simple linear layers for modal information extraction. (2) Previous works treat modal features as learnable embeddings, which causes the modal embeddings to gradually deviate from the original modal features during learning. We refer to this issue as Modal Information Forgetting. (3) Previous approaches fail to account for the significant differences in the distribution between behavior and modality, leading to the issue of representation misalignment. To address these challenges, this paper proposes a novel Dual REpresentAtion learning model for Multimodal Recommendation called DREAM. For sufficient information extraction, we introduce separate dual lines, including Behavior Line and Modal Line, in which the Modal-specific Encoder is applied to empower modal representations. To address the issue of Modal Information Forgetting, we introduce the Similarity Supervised Signal to constrain the modal representations. Additionally, we design a Behavior-Modal Alignment module to fuse the dual representations through Intra-Alignment and Inter-Alignment. Extensive experiments on three public datasets demonstrate that the proposed DREAM method achieves state-of-the-art (SOTA) results. The source code will be available upon acceptance.

cs.IR↗

arXiv2024

AlignRec: Aligning and Training in Multimodal Recommendations

With the development of multimedia systems, multimodal recommendations are playing an essential role, as they can leverage rich contexts beyond interactions. Existing methods mainly regard multimodal information as an auxiliary, using them to help learn ID features; However, there exist semantic gaps among multimodal content features and ID-based features, for which directly using multimodal information as an auxiliary would lead to misalignment in representations of users and items. In this paper, we first systematically investigate the misalignment issue in multimodal recommendations, and propose a solution named AlignRec. In AlignRec, the recommendation objective is decomposed into three alignments, namely alignment within contents, alignment between content and categorical ID, and alignment between users and items. Each alignment is characterized by a specific objective function and is integrated into our multimodal recommendation framework. To effectively train AlignRec, we propose starting from pre-training the first alignment to obtain unified multimodal features and subsequently training the following two alignments together with these features as input. As it is essential to analyze whether each multimodal feature helps in training and accelerate the iteration cycle of recommendation models, we design three new classes of metrics to evaluate intermediate performance. Our extensive experiments on three real-world datasets consistently verify the superiority of AlignRec compared to nine baselines. We also find that the multimodal features generated by AlignRec are better than currently used ones, which are to be open-sourced in our repository https://github.com/sjtulyf123/AlignRec_CIKM24.

cs.IR↗