arXiv subjects

Jianling Wang

Publications and source records attributed to Jianling Wang.

At least 19 recordsLinked to original sources

arXiv2025

Balancing Fine-tuning and RAG: A Hybrid Strategy for Dynamic LLM Recommendation Updates

Large Language Models (LLMs) empower recommendation systems through their advanced reasoning and planning capabilities. However, the dynamic nature of user interests and content poses a significant challenge: While initial fine-tuning aligns LLMs with domain knowledge and user preferences, it fails to capture such real-time changes, necessitating robust update mechanisms. This paper investigates strategies for updating LLM-powered recommenders, focusing on the trade-offs between ongoing fine-tuning and Retrieval-Augmented Generation (RAG). Using an LLM-powered user interest exploration system as a case study, we perform a comparative analysis of these methods across dimensions like cost, agility, and knowledge incorporation. We propose a hybrid update strategy that leverages the long-term knowledge adaptation of periodic fine-tuning with the agility of low-cost RAG. We demonstrate through live A/B experiments on a billion-user platform that this hybrid approach yields statistically significant improvements in user satisfaction, offering a practical and cost-effective framework for maintaining high-quality LLM-powered recommender systems.

cs.IR↗

arXiv2025

Serendipitous Recommendation with Multimodal LLM

Conventional recommendation systems succeed in identifying relevant content but often fail to provide users with surprising or novel items. Multimodal Large Language Models (MLLMs) possess the world knowledge and multimodal understanding needed for serendipity, but their integration into billion-item-scale platforms presents significant challenges. In this paper, we propose a novel hierarchical framework where fine-tuned MLLMs provide high-level guidance to conventional recommendation models, steering them towards more serendipitous suggestions. This approach leverages MLLM strengths in understanding multimodal content and user interests while retaining the efficiency of traditional models for item-level recommendation. This mitigates the complexity of applying MLLMs directly to vast action spaces. We also demonstrate a chain-of-thought strategy enabling MLLMs to discover novel user interests by first understanding video content and then identifying relevant yet unexplored interest clusters. Through live experiments within a commercial short-form video platform serving billions of users, we show that our MLLM-powered approach significantly improves both recommendation serendipity and user satisfaction.

cs.IR↗

arXiv2025

Flow Matching for Collaborative Filtering

Generative models have shown great promise in collaborative filtering by capturing the underlying distribution of user interests and preferences. However, existing approaches struggle with inaccurate posterior approximations and misalignment with the discrete nature of recommendation data, limiting their expressiveness and real-world performance. To address these limitations, we propose FlowCF, a novel flow-based recommendation system leveraging flow matching for collaborative filtering. We tailor flow matching to the unique challenges in recommendation through two key innovations: (1) a behavior-guided prior that aligns with user behavior patterns to handle the sparse and heterogeneous user-item interactions, and (2) a discrete flow framework to preserve the binary nature of implicit feedback while maintaining the benefits of flow matching, such as stable training and efficient inference. Extensive experiments demonstrate that FlowCF achieves state-of-the-art recommendation accuracy across various datasets with the fastest inference speed, making it a compelling approach for real-world recommender systems. The code is available at https://github.com/chengkai-liu/FlowCF.

cs.IR↗

arXiv2025

User Feedback Alignment for LLM-powered Exploration in Large-scale Recommendation Systems

Exploration, the act of broadening user experiences beyond their established preferences, is challenging in large-scale recommendation systems due to feedback loops and limited signals on user exploration patterns. Large Language Models (LLMs) offer potential solutions by leveraging their world knowledge to recommend novel content outside these loops. A key challenge is aligning LLMs with user preferences while preserving their knowledge and reasoning. To enhance planning for new user interests using LLMs, this paper introduces a novel approach that combines hierarchical planning with LLM inference-time scaling. This method aims to improve recommendation relevancy without compromising novelty. We decouple novelty and user-alignment, training separate LLMs for each objective. We then scale up the novelty-focused LLM's inference and select the best-of-n predictions using the user-aligned LLM. Live experiments demonstrate efficacy, showing significant gains in both user satisfaction (measured by watch activity and active user counts) and exploration diversity.

cs.IR↗

arXiv2025

Young stars discovered in dwarf spheroidal galaxies confirm their recent infall into the Milky way

Recent observations from the ESA Gaia satellite and with the ESO VLT, have identified the presence of a population of young, 0.5 to 2 Gyr old, stars in the halo and in dwarf spheroidal galaxies surrounding the Milky Way. It suggests that MW dwarf galaxies, currently devoid of gas, had, until recent times, enough gas to sustain a burst of star formation. The recent loss of gas coincides with their arrival in the vicinity of the Milky Way, in agreement with orbital predictions from Gaia that indicate that most dwarf galaxies reached the Milky Way halo less than 3 Gyr years ago. This completely changes the interpretation of their dynamics, mass, and dark matter content.

astro-ph.GA↗

arXiv2025

SGSST: Scaling Gaussian Splatting StyleTransfer

Applying style transfer to a full 3D environment is a challenging task that has seen many developments since the advent of neural rendering. 3D Gaussian splatting (3DGS) has recently pushed further many limits of neural rendering in terms of training speed and reconstruction quality. This work introduces SGSST: Scaling Gaussian Splatting Style Transfer, an optimization-based method to apply style transfer to pretrained 3DGS scenes. We demonstrate that a new multiscale loss based on global neural statistics, that we name SOS for Simultaneously Optimized Scales, enables style transfer to ultra-high resolution 3D scenes. Not only SGSST pioneers 3D scene style transfer at such high image resolutions, it also produces superior visual quality as assessed by thorough qualitative, quantitative and perceptual comparisons.

cs.CV↗

arXiv2025

Federated Conversational Recommender System

Conversational Recommender Systems (CRSs) have become increasingly popular as a powerful tool for providing personalized recommendation experiences. By directly engaging with users in a conversational manner to learn their current and fine-grained preferences, a CRS can quickly derive recommendations that are relevant and justifiable. However, existing conversational recommendation systems (CRSs) typically rely on a centralized training and deployment process, which involves collecting and storing explicitly-communicated user preferences in a centralized repository. These fine-grained user preferences are completely human-interpretable and can easily be used to infer sensitive information (e.g., financial status, political stands, and health information) about the user, if leaked or breached. To address the user privacy concerns in CRS, we first define a set of privacy protection guidelines for preserving user privacy under the conversational recommendation setting. Based on these guidelines, we propose a novel federated conversational recommendation framework that effectively reduces the risk of exposing user privacy by (i) de-centralizing both the historical interests estimation stage and the interactive preference elicitation stage and (ii) strictly bounding privacy leakage by enforcing user-level differential privacy with meticulously selected privacy budgets. Through extensive experiments, we show that the proposed framework not only satisfies these user privacy protection guidelines, but also enables the system to achieve competitive recommendation performance even when compared to the state-of-the-art non-private conversational recommendation approach.

cs.IR↗

arXiv2025

An intriguing coincidence between the majority of vast polar structure dwarfs and a recent major merger at the M31 position

A significant part of the Milky Way (MW) dwarf galaxies orbit within a Vast POlar Structure (VPOS), which is perpendicular to the Galactic disc and whose origin has not yet been identified. It includes the Large Magellanic Cloud (LMC) and its six dynamically associated dwarf galaxies. Andromeda Galaxy (M31) experienced a major merger two to three billion years ago, and its accurate modelling predicts that an associated tidal tail is pointing towards the Galaxy. Here, we tested a possible association between M31 tidal tail particles and MW dwarf galaxies, focusing first on the LMC and its associated dwarfs since they are less affected by ram pressure. We traced back these dwarf galaxy orbits by one billion years and calculated their association with the tidal tail particles in the 6D phase space, based on their proper motion from \textit{Gaia} DR3. We find that for low-mass MW models (total mass less than 5 $\times 10^{11} M_{\odot}$), the separation in the 6D space can be less than 1$σ$ for most of the M31 modelling, albeit with a significant degree of freedom due to the still unknown proper motion of M31. We further discover that many other dwarfs could also be associated with the M31 tidal tails if their motions had been radially slowed, as expected from the ram pressure exerted by the MW corona. This intriguing coincidence could explain the origin of the VPOS, which resulted from a matter exchange between M31 and MW.

astro-ph.GA↗

arXiv2025

Cold-Start Recommendation towards the Era of Large Language Models (LLMs): A Comprehensive Survey and Roadmap

Cold-start problem is one of the long-standing challenges in recommender systems, focusing on accurately modeling new or interaction-limited users or items to provide better recommendations. Due to the diversification of internet platforms and the exponential growth of users and items, the importance of cold-start recommendation (CSR) is becoming increasingly evident. At the same time, large language models (LLMs) have achieved tremendous success and possess strong capabilities in modeling user and item information, providing new potential for cold-start recommendations. However, the research community on CSR still lacks a comprehensive review and reflection in this field. Based on this, in this paper, we stand in the context of the era of large language models and provide a comprehensive review and discussion on the roadmap, related literature, and future directions of CSR. Specifically, we have conducted an exploration of the development path of how existing CSR utilizes information, from content features, graph relations, and domain information, to the world knowledge possessed by large language models, aiming to provide new insights for both the research and industrial communities on CSR. Related resources of cold-start recommendations are collected and continuously updated for the community in https://github.com/YuanchenBei/Awesome-Cold-Start-Recommendation.

cs.IR↗

arXiv2024

The accretion history of the Milky Way IV. Hints of recent star formation in Milky Way dwarf spheroidal galaxies

Dwarf spheroidal galaxies are known to be dominated by old stellar populations. This has led to the assumption that their gas-rich progenitors lost their gas during their infall in the Milky Way (MW) halo at distant look-back times. Here, we report a discovery of a tiny but robustly detected population of possibly young ($\sim$ 1 Gyr old) and intermediate-mass ($\rm 1.8 M_{\odot} \le M < 3 M_{\odot}$) stars in MW dwarf spheroidal galaxies. This was established on the basis of their positions in color-magnitude diagrams, after filtering out the bulk of the foreground MW using Gaia DR3 proper motions. We have considered the possibility that this population is made of evolved blue stragglers. For Sculptor, it seems unlikely, because 95.5% of its stars are older than 8 Gyr, leading to masses smaller than 0.9 M$_{\odot}$. This would only allow blue straggler masses of less than 1.8 M$_{\odot}$, which is much lower than what we observed. Alternatively, it would require the merger of three turnoff stars, which appears even more unlikely. On the other hand, the recent Gaia proper motion measurements of MW dwarf galaxies infer their low binding energies and large angular momenta, pointing to a more recent, $\le$ 3 Gyr, infall. Although the nature of the newly discovered stars still needs further confirmation, we find that they are consistent with the recent infall of the dwarf galaxies into the MW halo, when star formation occurred from the ram pressurization of their gas content before its removal by the hot Galactic corona. The abundance of this plausibly young population of stars is similar to the expectations drawn from hydrodynamical simulations. These results point to a novel origin for MW dwarf spheroidal galaxies.

astro-ph.GA↗

arXiv2024

TwinCL: A Twin Graph Contrastive Learning Model for Collaborative Filtering

In the domain of recommendation and collaborative filtering, Graph Contrastive Learning (GCL) has become an influential approach. Nevertheless, the reasons for the effectiveness of contrastive learning are still not well understood. In this paper, we challenge the conventional use of random augmentations on graph structure or embedding space in GCL, which may disrupt the structural and semantic information inherent in Graph Neural Networks. Moreover, fixed-rate data augmentation proves to be less effective compared to augmentation with an adaptive rate. In the initial training phases, significant perturbations are more suitable, while as the training approaches convergence, milder perturbations yield better results. We introduce a twin encoder in place of random augmentations, demonstrating the redundancy of traditional augmentation techniques. The twin encoder updating mechanism ensures the generation of more diverse contrastive views in the early stages, transitioning to views with greater similarity as training progresses. In addition, we investigate the learned representations from the perspective of alignment and uniformity on a hypersphere to optimize more efficiently. Our proposed Twin Graph Contrastive Learning model -- TwinCL -- aligns positive pairs of user and item embeddings and the representations from the twin encoder while maintaining the uniformity of the embeddings on the hypersphere. Our theoretical analysis and experimental results show that the proposed model optimizing alignment and uniformity with the twin encoder contributes to better recommendation accuracy and training efficiency performance. In comprehensive experiments on three public datasets, our proposed TwinCL achieves an average improvement of 5.6% (NDCG@10) in recommendation accuracy with faster training speed, while effectively mitigating popularity bias.

cs.IR↗

arXiv2024

Behavior-Dependent Linear Recurrent Units for Efficient Sequential Recommendation

Sequential recommender systems aims to predict the users' next interaction through user behavior modeling with various operators like RNNs and attentions. However, existing models generally fail to achieve the three golden principles for sequential recommendation simultaneously, i.e., training efficiency, low-cost inference, and strong performance. To this end, we propose RecBLR, an Efficient Sequential Recommendation Model based on Behavior-Dependent Linear Recurrent Units to accomplish the impossible triangle of the three principles. By incorporating gating mechanisms and behavior-dependent designs into linear recurrent units, our model significantly enhances user behavior modeling and recommendation performance. Furthermore, we unlock the parallelizable training as well as inference efficiency for our model by designing a hardware-aware scanning acceleration algorithm with a customized CUDA kernel. Extensive experiments on real-world datasets with varying lengths of user behavior sequences demonstrate RecBLR's remarkable effectiveness in simultaneously achieving all three golden principles - strong recommendation performance, training efficiency, and low-cost inference, while exhibiting excellent scalability to datasets with long user interaction histories.

cs.IR↗

arXiv2024

Mamba4Rec: Towards Efficient Sequential Recommendation with Selective State Space Models

Sequential recommendation aims to estimate the dynamic user preferences and sequential dependencies among historical user behaviors. Although Transformer-based models have proven to be effective for sequential recommendation, they suffer from the inference inefficiency problem stemming from the quadratic computational complexity of attention operators, especially for long behavior sequences. Inspired by the recent success of state space models (SSMs), we propose Mamba4Rec, which is the first work to explore the potential of selective SSMs for efficient sequential recommendation. Built upon the basic Mamba block which is a selective SSM with an efficient hardware-aware parallel algorithm, we design a series of sequential modeling techniques to further promote model performance while maintaining inference efficiency. Through experiments on public datasets, we demonstrate how Mamba4Rec effectively tackles the effectiveness-efficiency dilemma, outperforming both RNN- and attention-based baselines in terms of both effectiveness and efficiency. The code is available at https://github.com/chengkai-liu/Mamba4Rec.

cs.IR↗

arXiv2024

LLMs for User Interest Exploration in Large-scale Recommendation Systems

Traditional recommendation systems are subject to a strong feedback loop by learning from and reinforcing past user-item interactions, which in turn limits the discovery of novel user interests. To address this, we introduce a hybrid hierarchical framework combining Large Language Models (LLMs) and classic recommendation models for user interest exploration. The framework controls the interfacing between the LLMs and the classic recommendation models through "interest clusters", the granularity of which can be explicitly determined by algorithm designers. It recommends the next novel interests by first representing "interest clusters" using language, and employs a fine-tuned LLM to generate novel interest descriptions that are strictly within these predefined clusters. At the low level, it grounds these generated interests to an item-level policy by restricting classic recommendation models, in this case a transformer-based sequence recommender to return items that fall within the novel clusters generated at the high level. We showcase the efficacy of this approach on an industrial-scale commercial platform serving billions of users. Live experiments show a significant increase in both exploration of novel interests and overall user enjoyment of the platform.

cs.IR↗

arXiv2024

Empowering Large Language Models for Textual Data Augmentation

With the capabilities of understanding and executing natural language instructions, Large language models (LLMs) can potentially act as a powerful tool for textual data augmentation. However, the quality of augmented data depends heavily on the augmentation instructions provided, and the effectiveness can fluctuate across different downstream tasks. While manually crafting and selecting instructions can offer some improvement, this approach faces scalability and consistency issues in practice due to the diversity of downstream tasks. In this work, we address these limitations by proposing a new solution, which can automatically generate a large pool of augmentation instructions and select the most suitable task-informed instructions, thereby empowering LLMs to create high-quality augmented data for different downstream tasks. Empirically, the proposed approach consistently generates augmented data with better quality compared to non-LLM and LLM-based data augmentation methods, leading to the best performance on 26 few-shot learning tasks sourced from a wide range of application domains.

cs.CL↗

arXiv2024

Countering Mainstream Bias via End-to-End Adaptive Local Learning

Collaborative filtering (CF) based recommendations suffer from mainstream bias -- where mainstream users are favored over niche users, leading to poor recommendation quality for many long-tail users. In this paper, we identify two root causes of this mainstream bias: (i) discrepancy modeling, whereby CF algorithms focus on modeling mainstream users while neglecting niche users with unique preferences; and (ii) unsynchronized learning, where niche users require more training epochs than mainstream users to reach peak performance. Targeting these causes, we propose a novel end-To-end Adaptive Local Learning (TALL) framework to provide high-quality recommendations to both mainstream and niche users. TALL uses a loss-driven Mixture-of-Experts module to adaptively ensemble experts to provide customized local models for different users. Further, it contains an adaptive weight module to synchronize the learning paces of different users by dynamically adjusting weights in the loss. Extensive experiments demonstrate the state-of-the-art performance of the proposed model. Code and data are provided at \url{https://github.com/JP-25/end-To-end-Adaptive-Local-Leanring-TALL-}

cs.IR↗

arXiv2024

Large Language Models as Data Augmenters for Cold-Start Item Recommendation

The reasoning and generalization capabilities of LLMs can help us better understand user preferences and item characteristics, offering exciting prospects to enhance recommendation systems. Though effective while user-item interactions are abundant, conventional recommendation systems struggle to recommend cold-start items without historical interactions. To address this, we propose utilizing LLMs as data augmenters to bridge the knowledge gap on cold-start items during training. We employ LLMs to infer user preferences for cold-start items based on textual description of user historical behaviors and new item descriptions. The augmented training signals are then incorporated into learning the downstream recommendation models through an auxiliary pairwise loss. Through experiments on public Amazon datasets, we demonstrate that LLMs can effectively augment the training signals for cold-start items, leading to significant improvements in cold-start item recommendation for various recommendation models.

cs.IR↗

arXiv2023

The accretion history of the Milky Way. II. Internal kinematics of globular clusters and of dwarf galaxies

We study how structural properties of globular clusters and dwarf galaxies are linked to their orbits in the Milky Way halo. From the inner to the outer halo, orbital energy increases and stellar-systems gradually move out of internal equilibrium: in the inner halo, high-surface brightness globular clusters are at pseudo-equilibrium, while further away, low-surface brightness clusters and dwarfs appear more tidally disturbed. Dwarf galaxies are the latest to arrive into the halo as indicated by their large orbital energies and pericenters, and have no time for more than one orbit. Their (gas-rich) progenitors likely lost their gas during their recent arrival in the Galactic halo. If dwarfs are at equilibrium with their dark matter (DM) content, the DM density should anti-correlate with pericenter. However, the transformation of DM dominated dwarfs from gas-rich rotation-supported into gas-poor dispersion-supported systems is unlikely accomplished during a single orbit. We suggest instead that the above anti-correlation is brought by the combination of ram-pressure stripping and of Galactic tidal shocks. Recent gas removal leads to an expansion of their stellar content caused by the associated gravity loss, making them sufficiently fragile to be transformed near pericenter passage. Out of equilibrium dwarfs would explain the observed anti-correlation of kinematics-based DM density with pericenter without invoking DM density itself, questioning its previous estimates. Ram-pressure stripping and tidal shocks may contribute to the dwarf velocity dispersion excess. It predicts the presence of numerous stars in their outskirts and a few young stars in their cores.

astro-ph.GA↗