arXiv subjects

Wei Yuan

Publications and source records attributed to Wei Yuan.

At least 55 records · Page 3Linked to original sources

arXiv2025

Step-DeepResearch Technical Report

As LLMs shift toward autonomous agents, Deep Research has emerged as a pivotal metric. However, existing academic benchmarks like BrowseComp often fail to meet real-world demands for open-ended research, which requires robust skills in intent recognition, long-horizon decision-making, and cross-source verification. To address this, we introduce Step-DeepResearch, a cost-effective, end-to-end agent. We propose a Data Synthesis Strategy Based on Atomic Capabilities to reinforce planning and report writing, combined with a progressive training path from agentic mid-training to SFT and RL. Enhanced by a Checklist-style Judger, this approach significantly improves robustness. Furthermore, to bridge the evaluation gap in the Chinese domain, we establish ADR-Bench for realistic deep research scenarios. Experimental results show that Step-DeepResearch (32B) scores 61.4% on Scale AI Research Rubrics. On ADR-Bench, it significantly outperforms comparable models and rivals SOTA closed-source models like OpenAI and Gemini DeepResearch. These findings prove that refined training enables medium-sized models to achieve expert-level capabilities at industry-leading cost-efficiency.

cs.CL↗

arXiv2025

OneRec-V2 Technical Report

Recent breakthroughs in generative AI have transformed recommender systems through end-to-end generation. OneRec reformulates recommendation as an autoregressive generation task, achieving high Model FLOPs Utilization. While OneRec-V1 has shown significant empirical success in real-world deployment, two critical challenges hinder its scalability and performance: (1) inefficient computational allocation where 97.66% of resources are consumed by sequence encoding rather than generation, and (2) limitations in reinforcement learning relying solely on reward models. To address these challenges, we propose OneRec-V2, featuring: (1) Lazy Decoder-Only Architecture: Eliminates encoder bottlenecks, reducing total computation by 94% and training resources by 90%, enabling successful scaling to 8B parameters. (2) Preference Alignment with Real-World User Interactions: Incorporates Duration-Aware Reward Shaping and Adaptive Ratio Clipping to better align with user preferences using real-world feedback. Extensive A/B tests on Kuaishou demonstrate OneRec-V2's effectiveness, improving App Stay Time by 0.467%/0.741% while balancing multi-objective recommendations. This work advances generative recommendation scalability and alignment with real-world feedback, representing a step forward in the development of end-to-end recommender systems.

cs.IR↗

arXiv2025

OneRec Technical Report

Recommender systems have been widely used in various large-scale user-oriented platforms for many years. However, compared to the rapid developments in the AI community, recommendation systems have not achieved a breakthrough in recent years. For instance, they still rely on a multi-stage cascaded architecture rather than an end-to-end approach, leading to computational fragmentation and optimization inconsistencies, and hindering the effective application of key breakthrough technologies from the AI community in recommendation scenarios. To address these issues, we propose OneRec, which reshapes the recommendation system through an end-to-end generative approach and achieves promising results. Firstly, we have enhanced the computational FLOPs of the current recommendation model by 10 $\times$ and have identified the scaling laws for recommendations within certain boundaries. Secondly, reinforcement learning techniques, previously difficult to apply for optimizing recommendations, show significant potential in this framework. Lastly, through infrastructure optimizations, we have achieved 23.7% and 28.8% Model FLOPs Utilization (MFU) on flagship GPUs during training and inference, respectively, aligning closely with the LLM community. This architecture significantly reduces communication and storage overhead, resulting in operating expense that is only 10.6% of traditional recommendation pipelines. Deployed in Kuaishou/Kuaishou Lite APP, it handles 25% of total queries per second, enhancing overall App Stay Time by 0.54% and 1.24%, respectively. Additionally, we have observed significant increases in metrics such as 7-day Lifetime, which is a crucial indicator of recommendation experience. We also provide practical lessons and insights derived from developing, optimizing, and maintaining a production-scale recommendation system with significant real-world impact.

cs.IR↗

arXiv2025

Kwai Keye-VL 1.5 Technical Report

In recent years, the development of Large Language Models (LLMs) has significantly advanced, extending their capabilities to multimodal tasks through Multimodal Large Language Models (MLLMs). However, video understanding remains a challenging area due to the dynamic and information-dense nature of videos. Existing models struggle with the trade-off between spatial resolution and temporal coverage when processing video content. We present Keye-VL-1.5, which addresses fundamental challenges in video comprehension through three key innovations. First, we introduce a novel Slow-Fast video encoding strategy that dynamically allocates computational resources based on inter-frame similarity, processing key frames with significant visual changes at higher resolution (Slow pathway) while handling relatively static frames with increased temporal coverage at lower resolution (Fast pathway). Second, we implement a progressive four-stage pre-training methodology that systematically extends the model's context length from 8K to 128K tokens, enabling processing of longer videos and more complex visual content. Third, we develop a comprehensive post-training pipeline focusing on reasoning enhancement and human preference alignment, incorporating a 5-step chain-of-thought data construction process, iterative GSPO-based reinforcement learning with progressive prompt hinting for difficult cases, and alignment training. Through extensive evaluation on public benchmarks and rigorous internal human assessment, Keye-VL-1.5 demonstrates significant improvements over existing models, particularly excelling in video understanding tasks while maintaining competitive performance on general multimodal benchmarks.

cs.CV↗

arXiv2025

Meissner Effect and Nonreciprocal Charge Transport in Non-Topological 1T-CrTe2/FeTe Heterostructures

Interface-induced superconductivity has recently been achieved by stacking a magnetic topological insulator layer on an antiferromagnetic FeTe layer. However, the mechanism driving this emergent superconductivity remains unclear. Here, we employ molecular beam epitaxy to grow a 1T-CrTe2 layer, a two-dimensional ferromagnet with a Curie temperature up to room temperature, on a FeTe layer. These 1T-CrTe2/FeTe heterostructures show superconductivity with a critical temperature of ~12 K. Through magnetic force microscopy measurements, we observe the Meissner effect on the surface of the 1T-CrTe2 layer. Our electrical transport measurements reveal that the 1T-CrTe2/FeTe heterostructures exhibit nonreciprocal charge transport behavior, characterized by a large magneto-chiral anisotropy coefficient. The enhanced nonreciprocal charge transport in 1T-CrTe2/FeTe heterostructures provides a promising platform for exploring the magnetically controllable superconducting diode effect.

cond-mat.supr-con↗

arXiv2025

Towards Propagation-aware Representation Learning for Supervised Social Media Graph Analytics

Social media platforms generate vast, complex graph-structured data, facilitating diverse tasks such as rumor detection, bot identification, and influence modeling. Real-world applications like public opinion monitoring and stock trading -- which have a strong attachment to social media -- demand models that are performant across diverse tasks and datasets. However, most existing solutions are purely data-driven, exhibiting vulnerability to the inherent noise within social media data. Moreover, the reliance on task-specific model design challenges efficient reuse of the same model architecture on different tasks, incurring repetitive engineering efforts. To address these challenges in social media graph analytics, we propose a general representation learning framework that integrates a dual-encoder structure with a kinetic-guided propagation module. In addition to jointly modeling structural and contextual information with two encoders, our framework innovatively captures the information propagation dynamics within social media graphs by integrating principled kinetic knowledge. By deriving a propagation-aware encoder and corresponding optimization objective from a Markov chain-based transmission model, the representation learning pipeline receives a boost in its robustness to noisy data and versatility in diverse tasks. Extensive experiments verify that our approach achieves state-of-the-art performance with a unified architecture on a variety of social media graph mining tasks spanning graph classification, node classification, and link prediction. Besides, our solution exhibits strong zero-shot and few-shot transferability across datasets, demonstrating practicality when handling data-scarce tasks.

cs.SI↗

arXiv2025

Proxy Model-Guided Reinforcement Learning for Client Selection in Federated Recommendation

Federated recommender systems have emerged as a promising privacy-preserving paradigm, enabling personalized recommendation services without exposing users' raw data. By keeping data local and relying on a central server to coordinate training across distributed clients, FedRSs protect user privacy while collaboratively learning global models. However, most existing FedRS frameworks adopt fully random client selection strategy in each training round, overlooking the statistical heterogeneity of user data arising from diverse preferences and behavior patterns, thereby resulting in suboptimal model performance. While some client selection strategies have been proposed in the broader federated learning literature, these methods are typically designed for generic tasks and fail to address the unique challenges of recommendation scenarios, such as expensive contribution evaluation due to the large number of clients, and sparse updates resulting from long-tail item distributions. To bridge this gap, we propose ProxyRL-FRS, a proxy model-guided reinforcement learning framework tailored for client selection in federated recommendation. Specifically, we first introduce ProxyNCF, a dual-branch model deployed on each client, which augments standard Neural Collaborative Filtering with an additional proxy model branch that provides lightweight contribution estimation, thus eliminating the need for expensive per-round local training traditionally required to evaluate a client's contribution. Furthermore, we design a staleness-aware SA reinforcement learning agent that selects clients based on the proxy-estimated contribution, and is guided by a reward function balancing recommendation accuracy and embedding staleness, thereby enriching the update coverage of item embeddings. Experiments conducted on public recommendation datasets demonstrate the effectiveness of ProxyRL-FRS.

cs.IR↗

arXiv2025

On-Device Recommender Systems: A Comprehensive Survey

Recommender systems have been widely deployed in various real-world applications to help users identify content of interest from massive amounts of information. Traditional recommender systems work by collecting user-item interaction data in a cloud-based data center and training a centralized model to perform the recommendation service. However, such cloud-based recommender systems (CloudRSs) inevitably suffer from excessive resource consumption, response latency, as well as privacy and security risks concerning both data and models. Recently, driven by the advances in storage, communication, and computation capabilities of edge devices, there has been a shift of focus from CloudRSs to on-device recommender systems (DeviceRSs), which leverage the capabilities of edge devices to minimize centralized data storage requirements, reduce the response latency caused by communication overheads, and enhance user privacy and security by localizing data processing and model training. Despite the rapid rise of DeviceRSs, there is a clear absence of timely literature reviews that systematically introduce, categorize and contrast these methods. To bridge this gap, we aim to provide a comprehensive survey of DeviceRSs, covering three main aspects: (1) the deployment and inference of DeviceRSs (2) the training and update of DeviceRSs (3) the security and privacy of DeviceRSs. Furthermore, we provide a fine-grained and systematic taxonomy of the methods involved in each aspect, followed by a discussion regarding challenges and future research directions. This is the first comprehensive survey on DeviceRSs that covers a spectrum of tasks to fit various needs. We believe this survey will help readers effectively grasp the current research status in this field, equip them with relevant technical foundations, and stimulate new research ideas for developing DeviceRSs.

cs.IR↗

arXiv2025

Controllable and Stealthy Shilling Attacks via Dispersive Latent Diffusion

Recommender systems (RSs) are now fundamental to various online platforms, but their dependence on user-contributed data leaves them vulnerable to shilling attacks that can manipulate item rankings by injecting fake users. Although widely studied, most existing attack models fail to meet two critical objectives simultaneously: achieving strong adversarial promotion of target items while maintaining realistic behavior to evade detection. As a result, the true severity of shilling threats that manage to reconcile the two objectives remains underappreciated. To expose this overlooked vulnerability, we present DLDA, a diffusion-based attack framework that can generate highly effective yet indistinguishable fake users by enabling fine-grained control over target promotion. Specifically, DLDA operates in a pre-aligned collaborative embedding space, where it employs a conditional latent diffusion process to iteratively synthesize fake user profiles with precise target item control. To evade detection, DLDA introduces a dispersive regularization mechanism that promotes variability and realism in generated behavioral patterns. Extensive experiments on three real-world datasets and five popular RS models demonstrate that, compared to prior attacks, DLDA consistently achieves stronger item promotion while remaining harder to detect. These results highlight that modern RSs are more vulnerable than previously recognized, underscoring the urgent need for more robust defenses.

cs.LG↗

arXiv2025

A Provably Secure Network Protocol for Private Communication with Analysis and Tracing Resistance

Anonymous communication networks have emerged as crucial tools for obfuscating communication pathways and concealing user identities. However, their practical deployments face significant challenges, including susceptibility to artificial intelligence (AI)-powered metadata analysis, difficulties in decentralized architectures, and the absence of provable security guarantees. To address these issues, this paper proposes a novel decentralized anonymous routing protocol with resistance to tracing and traffic analysis. The protocol eliminates dependencies on the threshold model and trusted third-party setups, ensuring indistinguishable identity privacy even in highly adversarial environments. Different from traditional empirical security analysis of anonymous networks, this paper rigorously proves indistinguishable identity privacy for users even in extremely adversarial environments. Furthermore, simulations confirm its practical feasibility, demonstrating both security and efficiency. By achieving information sharing with privacy preservation, the proposed protocol offers a provably secure solution for privacy-preserving communication in digital environments.

cs.CR↗

arXiv2025

Watermarking Large Language Model-based Time Series Forecasting

Large Language Model-based Time Series Forecasting (LLMTS) has shown remarkable promise in handling complex and diverse temporal data, representing a significant step toward foundation models for time series analysis. However, this emerging paradigm introduces two critical challenges. First, the substantial commercial potential and resource-intensive development raise urgent concerns about intellectual property (IP) protection. Second, their powerful time series forecasting capabilities may be misused to produce misleading or fabricated deepfake time series data. To address these concerns, we explore watermarking the outputs of LLMTS models, that is, embedding imperceptible signals into the generated time series data that remain detectable by specialized algorithms. We propose a novel post-hoc watermarking framework, Waltz, which is broadly compatible with existing LLMTS models. Waltz is inspired by the empirical observation that time series patch embeddings are rarely aligned with a specific set of LLM tokens, which we term ``cold tokens''. Leveraging this insight, Waltz embeds watermarks by rewiring the similarity statistics between patch embeddings and cold token embeddings, and detects watermarks using similarity z-scores. To minimize potential side effects, we introduce a similarity-based embedding position identification strategy and employ projected gradient descent to constrain the watermark noise within a defined boundary. Extensive experiments using two popular LLMTS models across seven benchmark datasets demonstrate that Waltz achieves high watermark detection accuracy with minimal impact on the quality of the generated time series.

cs.IR↗

arXiv2025

Kwai Keye-VL Technical Report

While Multimodal Large Language Models (MLLMs) demonstrate remarkable capabilities on static images, they often fall short in comprehending dynamic, information-dense short-form videos, a dominant medium in today's digital landscape. To bridge this gap, we introduce \textbf{Kwai Keye-VL}, an 8-billion-parameter multimodal foundation model engineered for leading-edge performance in short-video understanding while maintaining robust general-purpose vision-language abilities. The development of Keye-VL rests on two core pillars: a massive, high-quality dataset exceeding 600 billion tokens with a strong emphasis on video, and an innovative training recipe. This recipe features a four-stage pre-training process for solid vision-language alignment, followed by a meticulous two-phase post-training process. The first post-training stage enhances foundational capabilities like instruction following, while the second phase focuses on stimulating advanced reasoning. In this second phase, a key innovation is our five-mode ``cold-start'' data mixture, which includes ``thinking'', ``non-thinking'', ``auto-think'', ``think with image'', and high-quality video data. This mixture teaches the model to decide when and how to reason. Subsequent reinforcement learning (RL) and alignment steps further enhance these reasoning capabilities and correct abnormal model behaviors, such as repetitive outputs. To validate our approach, we conduct extensive evaluations, showing that Keye-VL achieves state-of-the-art results on public video benchmarks and remains highly competitive on general image-based tasks (Figure 1). Furthermore, we develop and release the \textbf{KC-MMBench}, a new benchmark tailored for real-world short-video scenarios, where Keye-VL shows a significant advantage.

cs.CV↗

arXiv2025

Disorder-Induced Suppression of Antiferromagnetic Magnon Transport in Cr2O3

We explore the impact of spin disorder associated with structural defects on antiferromagnetic magnon transport by probing the spin-flop transition of Cr2O3 using spin Seebeck effect measurements. By fabricating homoepitaxial Cr2O3 films grown on smooth Cr2O3 crystals, we systematically vary the thickness of the films in which the presence of point defects modulates the spin-flop transition field. We find that magnon propagation through the film is entirely suppressed for film thickness greater than 3 nm, revealing the pivotal role of disorder in governing antiferromagnetic magnon transport.

cond-mat.mtrl-sci↗

arXiv2025

Lightweight Embeddings with Graph Rewiring for Collaborative Filtering

As recommendation services scale rapidly and their deployment now commonly involves resource-constrained edge devices, GNN-based recommender systems face significant challenges, including high embedding storage costs and runtime latency from graph propagations. Our previous work, LEGCF, effectively reduced embedding storage costs but struggled to maintain recommendation performance under stricter storage limits. Additionally, LEGCF did not address the extensive runtime computation costs associated with graph propagation, which involves heavy multiplication and accumulation operations (MACs). These challenges consequently hinder effective training and inference on resource-constrained edge devices. To address these limitations, we propose Lightweight Embeddings with Rewired Graph for Graph Collaborative Filtering (LERG), an improved extension of LEGCF. LERG retains LEGCFs compositional codebook structure but introduces quantization techniques to reduce the storage cost, enabling the inclusion of more meta-embeddings within the same storage. To optimize graph propagation, we pretrain the quantized compositional embedding table using the full interaction graph on resource-rich servers, after which a fine-tuning stage is engaged to identify and prune low-contribution entities via a gradient-free binary integer programming approach, constructing a rewired graph that excludes these entities (i.e., user/item nodes) from propagating signals. The quantized compositional embedding table with selective embedding participation and sparse rewired graph are transferred to edge devices which significantly reduce computation memory and inference time. Experiments on three public benchmark datasets, including an industry-scale dataset, demonstrate that LERG achieves superior recommendation performance while dramatically reducing storage and computation costs for graph-based recommendation services.

cs.IR↗

arXiv2025

Semiotic Reconstruction of Destination Expectation Constructs An LLM-Driven Computational Paradigm for Social Media Tourism Analytics

Social media's rise establishes user-generated content (UGC) as pivotal for travel decisions, yet analytical methods lack scalability. This study introduces a dual-method LLM framework: unsupervised expectation extraction from UGC paired with survey-informed supervised fine-tuning. Findings reveal leisure/social expectations drive engagement more than foundational natural/emotional factors. By establishing LLMs as precision tools for expectation quantification, we advance tourism analytics methodology and propose targeted strategies for experience personalization and social travel promotion. The framework's adaptability extends to consumer behavior research, demonstrating computational social science's transformative potential in marketing optimization.

cs.CL↗

arXiv2025

Data Watermarking for Sequential Recommender Systems

In the era of large foundation models, data has become a crucial component in building high-performance AI systems. As the demand for high-quality and large-scale data continues to rise, data copyright protection is attracting increasing attention. In this work, we explore the problem of data watermarking for sequential recommender systems, where a watermark is embedded into the target dataset and can be detected in models trained on that dataset. We focus on two settings: dataset watermarking, which protects the ownership of the entire dataset, and user watermarking, which safeguards the data of individual users. We present a method named Dataset Watermarking for Recommender Systems (DWRS) to address them. We define the watermark as a sequence of consecutive items inserted into normal users' interaction sequences. We define a Receptive Field (RF) to guide the inserting process to facilitate the memorization of the watermark. Extensive experiments on five representative sequential recommendation models and three benchmark datasets demonstrate the effectiveness of DWRS in protecting data copyright while preserving model utility.

cs.IR↗

arXiv2025

HMamba: Hyperbolic Mamba for Sequential Recommendation

Sequential recommendation systems have become a cornerstone of personalized services, adept at modeling the temporal evolution of user preferences by capturing dynamic interaction sequences. Existing approaches predominantly rely on traditional models, including RNNs and Transformers. Despite their success in local pattern recognition, Transformer-based methods suffer from quadratic computational complexity and a tendency toward superficial attention patterns, limiting their ability to infer enduring preference hierarchies in sequential recommendation data. Recent advances in Mamba-based sequential models introduce linear-time efficiency but remain constrained by Euclidean geometry, failing to leverage the intrinsic hyperbolic structure of recommendation data. To bridge this gap, we propose Hyperbolic Mamba, a novel architecture that unifies the efficiency of Mamba's selective state space mechanism with hyperbolic geometry's hierarchical representational power. Our framework introduces (1) a hyperbolic selective state space that maintains curvature-aware sequence modeling and (2) stabilized Riemannian operations to enable scalable training. Experiments across four benchmarks demonstrate that Hyperbolic Mamba achieves 3-11% improvement while retaining Mamba's linear-time efficiency, enabling real-world deployment. This work establishes a new paradigm for efficient, hierarchy-aware sequential modeling.

cs.IR↗

arXiv2025

LLM-based Bi-level Multi-interest Learning Framework for Sequential Recommendation

Sequential recommendation (SR) leverages users' dynamic preferences, with recent advances incorporating multi-interest learning to model diverse user interests. However, most multi-interest SR models rely on noisy, sparse implicit feedback, limiting recommendation accuracy. Large language models (LLMs) offer robust reasoning on low-quality data but face high computational costs and latency challenges for SR integration. We propose a novel LLM-based multi-interest SR framework combining implicit behavioral and explicit semantic perspectives. It includes two modules: the Implicit Behavioral Interest Module (IBIM), which learns from user behavior using a traditional SR model, and the Explicit Semantic Interest Module (ESIM), which uses clustering and prompt-engineered LLMs to extract semantic multi-interest representations from informative samples. Semantic insights from ESIM enhance IBIM's behavioral representations via modality alignment and semantic prediction tasks. During inference, only IBIM is used, ensuring efficient, LLM-free recommendations. Experiments on four real-world datasets validate the framework's effectiveness and practicality.

cs.IR↗