arXiv subjects

Hongyu Lu

Publications and source records attributed to Hongyu Lu.

At least 37 records · Page 2Linked to original sources

arXiv2025

Spectra of Magnetoroton and Chiral Graviton Modes of Fractional Chern Insulator

Employing the state-of-the-art time-dependent variational principle (TDVP) algorithm, we compute the spectra of charge-neutral excitations in the $ν=1/2$ (bosonic) \updated{ and $1/3$ (fermionic) fractional Chern insulator (FCI)} on the Haldane honeycomb lattice model. The magnetoroton visualized from the dynamic density structure factor acquires a minimum gap at finite momentum that can go soft with increasing interaction and give rise to a charge density wave (CDW) at the same wavevector. As the system approaches the FCI-to-CDW transition point, we observe a pronounced sharpening of the roton mode, suggesting that the magnetoroton behaves more like a quasiparticle as it softens. Notably, this occurs while the single-particle gap remains finite. Besides the magnetoroton at finite momentum, we also construct quadrupolar chiral operators in a discrete lattice and resolve the chiral graviton mode around the $Γ$ point of the Brillouin zone. Furthermore, we show the different chiralities of the gravitons of FCIs with opposite-sign Hall conductance for the first time.

cond-mat.str-el↗

arXiv2025

Universal Item Tokenization for Transferable Generative Recommendation

Recently, generative recommendation has emerged as a promising paradigm, attracting significant research attention. The basic framework involves an item tokenizer, which represents each item as a sequence of codes serving as its identifier, and a generative recommender that predicts the next item by autoregressively generating the target item identifier. However, in existing methods, both the tokenizer and the recommender are typically domain-specific, limiting their ability for effective transfer or adaptation to new domains. To this end, we propose UTGRec, a Universal item Tokenization approach for transferable Generative Recommendation. Specifically, we design a universal item tokenizer for encoding rich item semantics by adapting a multimodal large language model (MLLM). By devising tree-structured codebooks, we discretize content representations into corresponding codes for item tokenization. To effectively learn the universal item tokenizer on multiple domains, we introduce two key techniques in our approach. For raw content reconstruction, we employ dual lightweight decoders to reconstruct item text and images from discrete representations to capture general knowledge embedded in the content. For collaborative knowledge integration, we assume that co-occurring items are similar and integrate collaborative signals through co-occurrence alignment and reconstruction. Finally, we present a joint learning framework to pre-train and adapt the transferable generative recommender across multiple domains. Extensive experiments on four public datasets demonstrate the superiority of UTGRec compared to both traditional and generative recommendation baselines.

cs.IR↗

arXiv2025

Continuous Transition between Bosonic Fractional Chern Insulator and Superfluid

The properties of fractional Chern insulator (FCI) phases and the phase transitions between FCI and Mott insulators (MI) in bosonic systems are well studied. The continuous transitions between FCI and superfluid (SF), however, despite the inspiring field theoretical predictions, have not been directly verified. The existing numerical results of the FCI-SF transition are either indirect or clearly first-order. Here, by simply tuning the bandwidth of the Haldane honeycomb lattice model, we find direct transitions from a bosonic FCI at $ν=1/2$ filling of a flat Chern band to two SF states with bosons condensed at momenta M or $Γ$, respectively. While the FCI-SF(M) transition is first-order, the FCI-SF($Γ$) transition is found continuous, and the bipartite entanglement entropy at the critical point with the area-law scaling is consistent with the critical theories. Through finite size criticality analysis, the obtained critical exponents $β\approx$ 0.35(5) and $ν\approx$ 0.62(12) are both compatible with those of the 3D XY universality class within numerical uncertainty and possibly more exotic beyond-Landau ones. This letter thence presents a direct numerical demonstration of a continuous FCI-SF transition between topologically ordered phase and spontaneous continuous symmetry-breaking phase, and further indicates the zero-field bosonic FCI might be realized from a SF state by gradually flattening the dispersion of the Chern band, through the (quasi)adiabatic preparation in ultracold atom systems.

cond-mat.str-el↗

arXiv2025

Slow Thinking for Sequential Recommendation

To develop effective sequential recommender systems, numerous methods have been proposed to model historical user behaviors. Despite the effectiveness, these methods share the same fast thinking paradigm. That is, for making recommendations, these methods typically encodes user historical interactions to obtain user representations and directly match these representations with candidate item representations. However, due to the limited capacity of traditional lightweight recommendation models, this one-step inference paradigm often leads to suboptimal performance. To tackle this issue, we present a novel slow thinking recommendation model, named STREAM-Rec. Our approach is capable of analyzing historical user behavior, generating a multi-step, deliberative reasoning process, and ultimately delivering personalized recommendations. In particular, we focus on two key challenges: (1) identifying the suitable reasoning patterns in recommender systems, and (2) exploring how to effectively stimulate the reasoning capabilities of traditional recommenders. To this end, we introduce a three-stage training framework. In the first stage, the model is pretrained on large-scale user behavior data to learn behavior patterns and capture long-range dependencies. In the second stage, we design an iterative inference algorithm to annotate suitable reasoning traces by progressively refining the model predictions. This annotated data is then used to fine-tune the model. Finally, in the third stage, we apply reinforcement learning to further enhance the model generalization ability. Extensive experiments validate the effectiveness of our proposed method.

cs.IR↗

arXiv2024

Enhancing Graph Contrastive Learning with Reliable and Informative Augmentation for Recommendation

Graph neural network(GNN) has been a powerful approach in collaborative filtering(CF) due to its ability to model high-order user-item relationships. Recently, to alleviate the data sparsity and enhance representation learning, many efforts have been conducted to integrate contrastive learning(CL) with GNNs. Despite the promising improvements, the contrastive view generation based on structure and representation perturbations in existing methods potentially disrupts the collaborative information in contrastive views, resulting in limited effectiveness of positive alignment. To overcome this issue, we propose CoGCL, a novel framework that aims to enhance graph contrastive learning by constructing contrastive views with stronger collaborative information via discrete codes. The core idea is to map users and items into discrete codes rich in collaborative information for reliable and informative contrastive view generation. To this end, we initially introduce a multi-level vector quantizer in an end-to-end manner to quantize user and item representations into discrete codes. Based on these discrete codes, we enhance the collaborative information of contrastive views by considering neighborhood structure and semantic relevance respectively. For neighborhood structure, we propose virtual neighbor augmentation by treating discrete codes as virtual neighbors, which expands an observed user-item interaction into multiple edges involving discrete codes. Regarding semantic relevance, we identify similar users/items based on shared discrete codes and interaction targets to generate the semantically relevant view. Through these strategies, we construct contrastive views with stronger collaborative information and develop a triple-view graph contrastive learning approach. Extensive experiments on four public datasets demonstrate the effectiveness of our proposed approach.

cs.IR↗

arXiv2024

From Fractional Quantum Anomalous Hall Smectics to Polar Smectic Metals: Nontrivial Interplay Between Electronic Liquid Crystal Order and Topological Order in Correlated Topological Flat Bands

Symmetry-breaking orders can not only compete with each other, but also be interwined, and the interwined topological and symmetry-breaking orders make the situation more intriguing. This work examines the archetypal correlated flat band model on a checkerboard lattice at filling $ν=2/3$ and we find the unique interplay between smectic charge order and topological order gives rise to two novel quantum states. As the interaction strength increases, the system first transitions from a Fermi liquid into FQAH smectic (FQAHS) state, where FQAH topological order coexists cooperatively with smectic charge order with enlarged ground-state degeneracy and interestingly, the Hall conductivity is $σ_{xy}=ν=2/3$, different from the band-folding or doping scenarios. Further increasing the interaction strength, the system undergoes another quantum phase transition and evolves into a polar smectic metal (PSM) state. This emergent PSM is an anisotropic non-Fermi liquid, whose interstripe tunneling is irrelevant while it is metallic inside each stripe. Different from the FQAHS and conventional smectic orders, this PSM spontaneously breaks the two-fold rotational symmetry, resulting in a nonzero electric dipole moment and ferroelectric order. In addition to the exotic ground states, large-scale numerical simulations are also used to study low-energy excitations and thermodynamic characteristics. We find the onset temperature of the incompressible FQAHS state, which also coincides with the onset of non-polar smectic order, is dictated by the magneto-roton modes. Above this onset temperature, the PSM state exists at intermediate-temperature regime. Although the T = 0 quantum phase transition between PSM and FQAHS is first order, the thermal FQAHS-PSM transition could be continuous. We expect the features of the exotic states and thermal phase transitions could be accessed in future experiments.

cond-mat.str-el↗

arXiv2024

A Framework for Assessing Cumulative Exposure to Extreme Temperatures During Transit Trip

The combined influence of urban heat islands, climate change, and extreme temperature events are increasingly impacting transit travelers, especially vulnerable populations such as older adults, people with disabilities, and those with chronic diseases. Previous studies have generally attempted to address this issue at either the micro- or macro-level, but each approach presents different limitations in modeling the impacts on transit trips. Other research proposes a meso-level approach to address some of these gaps, but the use of additive exposure calculation and spatial shortest path routing poses constraints meso-modeling accuracy. This study introduces HeatPath Analyzer, a framework to assess the exposure of transit riders to extreme temperatures, using TransitSim 4.0 to generate second-by-second spatio-temporal trip trajectories, the traveler activity profiles, and thermal comfort levels along the entire journey. The approach uses heat stress combines the standards proposed by the NWS and CDC to estimate cumulative exposure for transit riders, with specific parameters tailored to the elderly and people with disabilities. The framework assesses the influence of extreme heat and winter chill. A case study in Atlanta, GA, reveals that 10.2% of trips on an average summer weekday in 2019 were at risk of extreme heat. The results uncover exposure disparities across different transit trip mode segments, and across mitigation-based and adaptation-based strategies. While the mitigation-based strategy highlights high-exposure segments such as long ingress and egress, adaptation should be prioritized toward the middle or second half of the trip when a traveler is waiting for transit or transferring between routes. A comparison between the traditional additive approach and the dynamic approach presented also shows significant disparities, which, if overlooked, can mislead policy decisions.

stat.AP↗

arXiv2024

Transit Rider Heat Stress in Atlanta, GA under Current and Future Climate Scenarios

Transit is a crucial mode of transportation, especially in urban areas and for urban and rural disadvantaged communities. Because extreme temperatures often pose threats to the elderly, members of the disability community, and other vulnerable populations, this study seeks to understand the level of influence that extreme temperatures may have on transit users across different demographic groups. In this case study for Atlanta, GA, heat stress is predicted for 2019 transit riders (using transit rider activity survey data) and for three future climate scenarios, SSP245, SSP370, and SSP585, into the year 2100. The HeatPath Analyzer and TransitSim 4.0 models were applied to predict cumulative heat exposure and trip-level risk for 35,999 trip equivalents for an average Atlanta area weekday in the summer of 2019. The analyses show that under 2019 weather conditions, 8.33% of summer trips were estimated to be conducted under extreme heat. With the projected future climate conditions, the percentage of trips under extreme heat risk grows steadily. By 2100, 37.1%, 56.1%, and 76.4% are projected to be under extreme heat risk for scenarios SSP245, SSP370, and SSP585, respectively. Under current weather conditions, Atlanta transit riders that own no vehicles and transit riders that are African American are disproportionately influenced by extreme heat. The disparity between these two groups and other groups of transit riders becomes wider as climate change continues to exacerbate. The findings of the study highlight an urgent need to implement heat mitigation and adaptation strategies in urban transit networks.

cs.SI↗

arXiv2024

Linear Periodically Time-Variant Digital PLL Phase Noise Modeling Using Conversion Matrices and Uncorrelated Upsampling

This paper introduces a conversion matrix method for linear periodically time-variant (LPTV) digital phase-locked loop (DPLL) phase noise modeling that offers precise and computationally efficient results to enable rapid design iteration and optimization. Unlike many previous studies, which either assume linear time-invariance (LTI) and therefore overlook phase noise aliasing effects, or solve LPTV systems with noise folding and multiple sampling rate conversions that heightens modeling and computational complexity, the proposed conversion matrix method allows the designer to represent LPTV systems using intuitive LTI-like transfer functions with excellent accuracy. Additionally, the uncorrelated upsampling method addresses the cross-correlated spectrum of cyclostationary noise sources by a simple matrix multiplication. This eliminates the need to consider the beat frequency of the upsampled noise source and the system with different sampling rates, thus improving computational efficiency. The proposed algorithm is applied to modeling a integer-N DPLL with time-varying proportional loop gain, and the modeling accuracy is validated with Simulink transient simulations.

eess.SP↗

arXiv2024

Cross Domain LifeLong Sequential Modeling for Online Click-Through Rate Prediction

Deep neural networks (DNNs) that incorporated lifelong sequential modeling (LSM) have brought great success to recommendation systems in various social media platforms. While continuous improvements have been made in domain-specific LSM, limited work has been done in cross-domain LSM, which considers modeling of lifelong sequences of both target domain and source domain. In this paper, we propose Lifelong Cross Network (LCN) to incorporate cross-domain LSM to improve the click-through rate (CTR) prediction in the target domain. The proposed LCN contains a LifeLong Attention Pyramid (LAP) module that comprises of three levels of cascaded attentions to effectively extract interest representations with respect to the candidate item from lifelong sequences. We also propose Cross Representation Production (CRP) module to enforce additional supervision on the learning and alignment of cross-domain representations so that they can be better reused on learning of the CTR prediction in the target domain. We conducted extensive experiments on WeChat Channels industrial dataset as well as on benchmark dataset. Results have revealed that the proposed LCN outperforms existing work in terms of both prediction accuracy and online performance.

cs.IR↗

arXiv2024

Thermodynamic Response and Neutral Excitations in Integer and Fractional Quantum Anomalous Hall States Emerging from Correlated Flat Bands

Integer and fractional Chern insulators have been extensively explored in correlated flat band models. Recently, the prediction and experimental observation of fractional quantum anomalous Hall (FQAH) states with spontaneous time-reversal-symmetry breaking have garnered attention. While the thermodynamics of integer quantum anomalous Hall (IQAH) states have been systematically studied, our theoretical knowledge on thermodynamic properties of FQAH states has been severely limited. Here, we delve into the general thermodynamic response and collective excitations of both IQAH and FQAH states within the paradigmatic flat Chern-band model with remote band considered. Our key findings include: i) In both $ν$ = 1 IQAH and $ν$ = 1/3 FQAH states, even without spin fluctuations, the charge-neutral collective excitations would lower the onset temperature of these topological states, to a value significantly smaller than the charge gap, due to band-mixing and multi-particle scattering; ii) By employing large-scale thermodynamic simulations in FQAH states in the presence of strong inter-band mixing between C = $\pm1$ bands, we find that the lowest collective excitations manifest as the zero-momentum excitons in the IQAH state, whereas in the FQAH state, they take the form of magneto-rotons with finite momentum; iii) The unique charge oscillations in FQAH states are exhibited with distinct experimental signatures, which we propose to detect in future experiments.

cond-mat.str-el↗

arXiv2024

Adapting Large Language Models by Integrating Collaborative Semantics for Recommendation

Recently, large language models (LLMs) have shown great potential in recommender systems, either improving existing recommendation models or serving as the backbone. However, there exists a large semantic gap between LLMs and recommender systems, since items to be recommended are often indexed by discrete identifiers (item ID) out of the LLM's vocabulary. In essence, LLMs capture language semantics while recommender systems imply collaborative semantics, making it difficult to sufficiently leverage the model capacity of LLMs for recommendation. To address this challenge, in this paper, we propose a new LLM-based recommendation model called LC-Rec, which can better integrate language and collaborative semantics for recommender systems. Our approach can directly generate items from the entire item set for recommendation, without relying on candidate items. Specifically, we make two major contributions in our approach. For item indexing, we design a learning-based vector quantization method with uniform semantic mapping, which can assign meaningful and non-conflicting IDs (called item indices) for items. For alignment tuning, we propose a series of specially designed tuning tasks to enhance the integration of collaborative semantics in LLMs. Our fine-tuning tasks enforce LLMs to deeply integrate language and collaborative semantics (characterized by the learned item indices), so as to achieve an effective adaptation to recommender systems. Extensive experiments demonstrate the effectiveness of our method, showing that our approach can outperform a number of competitive baselines including traditional recommenders and existing LLM-based recommenders. Our code is available at https://github.com/RUCAIBox/LC-Rec/.

cs.IR↗

arXiv2024

Interaction-driven Roton Condensation in C = 2/3 Fractional Quantum Anomalous Hall State

The interplay of topological order and charge order exhibits rich physics. Recent experiments that succesfully realized the frational quantum anomalous Hall (FQAH) effect in twisted MoTe$_2$ bilayers and rhombohedral multilayer graphene without external magnetic field further call for deeper understanding of the relation between topological order and charge order in quantum moiré materials. In the archetypal correlated flat-band model on checkerboard lattice, a FQAH smectic state with coexistent topological order and smectic charge order has been numerically discovered at filling $ν$ = 2/3. In this work, we explore the global ground-state phase diagram of the model with competing interactions and find a C = 2/3 FQAH phase surrounded by four different charge density wave (CDW) phases. In particular, we identify a FQAH-CDW transition triggered by roton condensation, in that, the minimal roton gap continues to decrease at the same finite momentum, along with the diverging density flucuations at the transition point, after which the system enters into a CDW metal phase with the same ordered wavevector. Our discovery points out that the charge-neutral roton modes can play a significant role in a transition from FQAH topological order to CDW symmetry-breaking order, discussed in FQH literature while severely neglected in FQAH systems.

cond-mat.str-el↗

arXiv2024

Large Language Models are Zero-Shot Rankers for Recommender Systems

Recently, large language models (LLMs) (e.g., GPT-4) have demonstrated impressive general-purpose task-solving abilities, including the potential to approach recommendation tasks. Along this line of research, this work aims to investigate the capacity of LLMs that act as the ranking model for recommender systems. We first formalize the recommendation problem as a conditional ranking task, considering sequential interaction histories as conditions and the items retrieved by other candidate generation models as candidates. To solve the ranking task by LLMs, we carefully design the prompting template and conduct extensive experiments on two widely-used datasets. We show that LLMs have promising zero-shot ranking abilities but (1) struggle to perceive the order of historical interactions, and (2) can be biased by popularity or item positions in the prompts. We demonstrate that these issues can be alleviated using specially designed prompting and bootstrapping strategies. Equipped with these insights, zero-shot LLMs can even challenge conventional recommendation models when ranking candidates are retrieved by multiple candidate generators. The code and processed datasets are available at https://github.com/RUCAIBox/LLMRank.

cs.IR↗

arXiv2023

Scaling Law of Large Sequential Recommendation Models

Scaling of neural networks has recently shown great potential to improve the model capacity in various fields. Specifically, model performance has a power-law relationship with model size or data size, which provides important guidance for the development of large-scale models. However, there is still limited understanding on the scaling effect of user behavior models in recommender systems, where the unique data characteristics (e.g. data scarcity and sparsity) pose new challenges to explore the scaling effect in recommendation tasks. In this work, we focus on investigating the scaling laws in large sequential recommendation models. Specially, we consider a pure ID-based task formulation, where the interaction history of a user is formatted as a chronological sequence of item IDs. We don't incorporate any side information (e.g. item text), because we would like to explore how scaling law holds from the perspective of user behavior. With specially improved strategies, we scale up the model size to 0.8B parameters, making it feasible to explore the scaling effect in a diverse range of model sizes. As the major findings, we empirically show that scaling law still holds for these trained models, even in data-constrained scenarios. We then fit the curve for scaling law, and successfully predict the test loss of the two largest tested model scales. Furthermore, we examine the performance advantage of scaling effect on five challenging recommendation tasks, considering the unique issues (e.g. cold start, robustness, long-term preference) in recommender systems. We find that scaling up the model size can greatly boost the performance on these challenging tasks, which again verifies the benefits of large recommendation models.

cs.IR↗

arXiv2023

Ubiquitous nematic Dirac semimetal emerging from interacting quadratic band touching system

Quadratic band touching (QBT) points are widely observed in 2D and 3D materials, including bilayer graphene and Luttinger semimetals, and attract significant attention from theory to experiment. However, even in its simplest form, the 2D checkerboard lattice QBT model, the phase diagram characterized by temperature and interaction strength still remains unknown beyond the weak-coupling regime. Intense debates persist regarding the existence of various interaction-driven insulating states in this system. To address these uncertainties, we employ thermal tensor network simulations, specifically exponential tensor renormalization group and tangent space tensor renormalization group, along with density matrix renormalization group calculations to provide a comprehensive finite-temperature phase diagram for this model and shed light on previous ambiguities. Notably, our findings reveal the emergence of a robust bond-nematic Dirac semimetal phase with distinct thermodynamic properties that set it part from the nematic insulating state and other symmetry broken states. This previously overlooked feature is found to be ubiquitous in interacting QBT systems. We also discuss the implications of these results for experimental systems such as bilayer graphene and iridate compounds.

cond-mat.str-el↗

arXiv2023

Streaming CTR Prediction: Rethinking Recommendation Task for Real-World Streaming Data

The Click-Through Rate (CTR) prediction task is critical in industrial recommender systems, where models are usually deployed on dynamic streaming data in practical applications. Such streaming data in real-world recommender systems face many challenges, such as distribution shift, temporal non-stationarity, and systematic biases, which bring difficulties to the training and utilizing of recommendation models. However, most existing studies approach the CTR prediction as a classification task on static datasets, assuming that the train and test sets are independent and identically distributed (a.k.a, i.i.d. assumption). To bridge this gap, we formulate the CTR prediction problem in streaming scenarios as a Streaming CTR Prediction task. Accordingly, we propose dedicated benchmark settings and metrics to evaluate and analyze the performance of the models in streaming data. To better understand the differences compared to traditional CTR prediction tasks, we delve into the factors that may affect the model performance, such as parameter scale, normalization, regularization, etc. The results reveal the existence of the ''streaming learning dilemma'', whereby the same factor may have different effects on model performance in the static and streaming scenarios. Based on the findings, we propose two simple but inspiring methods (i.e., tuning key parameters and exemplar replay) that significantly improve the effectiveness of the CTR models in the new streaming scenario. We hope our work will inspire further research on streaming CTR prediction and help improve the robustness and adaptability of recommender systems.

cs.IR↗

arXiv2022

Interaction-driven quantum anomalous Hall insulator in Dirac semimetal

The interaction-driven quantum anomalous Hall (QAH) insulator has been sought for a long time in a Dirac semimetal with linear band touching points at the Fermi level. By combining exact diagonalization, density matrix renormalization group, and analytical methods, we study a spinless fermion system on the checkerboard lattice with two fold rotational symmetry, which realizes two Dirac band touching points in the absence of interaction. At weak coupling, the Dirac semimetal is stable. At a finite density-density repulsive interaction, we analyze possible symmetry broken states, and find that an QAH state is stabilized when the interaction strength exceeds the energy scale controlling the separation between the Dirac points. Through numerical simulations, we verify the existence of the QAH phase with spontaneous time-reversal symmetry breaking and quantized Chern number $C = 1$.

cond-mat.str-el↗