arXiv subjects

Yi Zhang

Publications and source records attributed to Yi Zhang.

At least 397 records · Page 22Linked to original sources

arXiv2025

Hierarchical Optimization via LLM-Guided Objective Evolution for Mobility-on-Demand Systems

Online ride-hailing platforms aim to deliver efficient mobility-on-demand services, often facing challenges in balancing dynamic and spatially heterogeneous supply and demand. Existing methods typically fall into two categories: reinforcement learning (RL) approaches, which suffer from data inefficiency, oversimplified modeling of real-world dynamics, and difficulty enforcing operational constraints; or decomposed online optimization methods, which rely on manually designed high-level objectives that lack awareness of low-level routing dynamics. To address this issue, we propose a novel hybrid framework that integrates large language model (LLM) with mathematical optimization in a dynamic hierarchical system: (1) it is training-free, removing the need for large-scale interaction data as in RL, and (2) it leverages LLM to bridge cognitive limitations caused by problem decomposition by adaptively generating high-level objectives. Within this framework, LLM serves as a meta-optimizer, producing semantic heuristics that guide a low-level optimizer responsible for constraint enforcement and real-time decision execution. These heuristics are refined through a closed-loop evolutionary process, driven by harmony search, which iteratively adapts the LLM prompts based on feasibility and performance feedback from the optimization layer. Extensive experiments based on scenarios derived from both the New York and Chicago taxi datasets demonstrate the effectiveness of our approach, achieving an average improvement of 16% compared to state-of-the-art baselines.

cs.AI↗

arXiv2025

Transfer Faster, Price Smarter: Minimax Dynamic Pricing under Cross-Market Preference Shift

We study contextual dynamic pricing when a target market can leverage K auxiliary markets -- offline logs or concurrent streams -- whose mean utilities differ by a structured preference shift. We propose Cross-Market Transfer Dynamic Pricing (CM-TDP), the first algorithm that provably handles such model-shift transfer and delivers minimax-optimal regret for both linear and non-parametric utility models. For linear utilities of dimension d, where the difference between source- and target-task coefficients is $s_{0}$-sparse, CM-TDP attains regret $\tilde{O}((d*K^{-1}+s_{0})\log T)$. For nonlinear demand residing in a reproducing kernel Hilbert space with effective dimension $α$, complexity $β$ and task-similarity parameter $H$, the regret becomes $\tilde{O}\!(K^{-2αβ/(2αβ+1)}T^{1/(2αβ+1)} + H^{2/(2α+1)}T^{1/(2α+1)})$, matching information-theoretic lower bounds up to logarithmic factors. The RKHS bound is the first of its kind for transfer pricing and is of independent interest. Extensive simulations show up to 50% lower cumulative regret and 5 times faster learning relative to single-market pricing baselines. By bridging transfer learning, robust aggregation, and revenue optimization, CM-TDP moves toward pricing systems that transfer faster, price smarter.

stat.ME↗

arXiv2025

Kernel Regression in Structured Non-IID Settings: Theory and Implications for Denoising Score Learning

Kernel ridge regression (KRR) is a foundational tool in machine learning, with recent work emphasizing its connections to neural networks. However, existing theory primarily addresses the i.i.d. setting, while real-world data often exhibits structured dependencies - particularly in applications like denoising score learning where multiple noisy observations derive from shared underlying signals. We present the first systematic study of KRR generalization for non-i.i.d. data with signal-noise causal structure, where observations represent different noisy views of common signals. By developing a novel blockwise decomposition method that enables precise concentration analysis for dependent data, we derive excess risk bounds for KRR that explicitly depend on: (1) the kernel spectrum, (2) causal structure parameters, and (3) sampling mechanisms (including relative sample sizes for signals and noises). We further apply our results to denoising score learning, establishing generalization guarantees and providing principled guidance for sampling noisy data points. This work advances KRR theory while providing practical tools for analyzing dependent data in modern machine learning applications.

stat.ML↗

arXiv2025

Qwen3Guard Technical Report

As large language models (LLMs) become more capable and widely used, ensuring the safety of their outputs is increasingly critical. Existing guardrail models, though useful in static evaluation settings, face two major limitations in real-world applications: (1) they typically output only binary "safe/unsafe" labels, which can be interpreted inconsistently across diverse safety policies, rendering them incapable of accommodating varying safety tolerances across domains; and (2) they require complete model outputs before performing safety checks, making them fundamentally incompatible with streaming LLM inference, thereby preventing timely intervention during generation and increasing exposure to harmful partial outputs. To address these challenges, we present Qwen3Guard, a series of multilingual safety guardrail models with two specialized variants: Generative Qwen3Guard, which casts safety classification as an instruction-following task to enable fine-grained tri-class judgments (safe, controversial, unsafe); and Stream Qwen3Guard, which introduces a token-level classification head for real-time safety monitoring during incremental text generation. Both variants are available in three sizes (0.6B, 4B, and 8B parameters) and support up to 119 languages and dialects, providing comprehensive, scalable, and low-latency safety moderation for global LLM deployments. Evaluated across English, Chinese, and multilingual benchmarks, Qwen3Guard achieves state-of-the-art performance in both prompt and response safety classification. All models are released under the Apache 2.0 license for public use.

cs.CL↗

arXiv2025

Beyond Static LLM Policies: Imitation-Enhanced Reinforcement Learning for Recommendation

Recommender systems (RecSys) have become critical tools for enhancing user engagement by delivering personalized content across diverse digital platforms. Recent advancements in large language models (LLMs) demonstrate significant potential for improving RecSys, primarily due to their exceptional generalization capabilities and sophisticated contextual understanding, which facilitate the generation of flexible and interpretable recommendations. However, the direct deployment of LLMs as primary recommendation policies presents notable challenges, including persistent latency issues stemming from frequent API calls and inherent model limitations such as hallucinations and biases. To address these issues, this paper proposes a novel offline reinforcement learning (RL) framework that leverages imitation learning from LLM-generated trajectories. Specifically, inverse reinforcement learning is employed to extract robust reward models from LLM demonstrations. This approach negates the need for LLM fine-tuning, thereby substantially reducing computational overhead. Simultaneously, the RL policy is guided by the cumulative rewards derived from these demonstrations, effectively transferring the semantic insights captured by the LLM. Comprehensive experiments conducted on two benchmark datasets validate the effectiveness of the proposed method, demonstrating superior performance when compared against state-of-the-art RL-based and in-context learning baselines. The code can be found at https://github.com/ArronDZhang/IL-Rec.

cs.IR↗

arXiv2025

Cocycle weighted infinitesimal bialgebras and pre-Lie algebras on rooted trees

The concept of weighted infinitesimal bialgebras provides an algebraic framework for understanding the non-homogeneous associative Yang-Baxter equation. In this paper, we endow the space of decorated planar rooted forests with a two-parameters family of coproducts, making it into a weighted infinitesimal bialgebra. A combinatorial characterization of the coproducts is given via the notion of forest biideals. Furthermore, by constructing a bilinear symmetric form and introducing a new grafting operation on rooted forests, we describe the associated dual products. We also introduce the notion of the pair-weight 1-cocycle condition and investigate the universal properties of decorated planar rooted forests satisfying this condition. This leads to the definition of a weighted $Ω$-cocycle infinitesimal unitary bialgebra. As applications, we identify the initial object in the category of free cocycle infinitesimal unitary bialgebras on undecorated planar rooted forests, corresponding to the well-known noncommutative Connes-Kreimer Hopf algebra. In addition, we establish isomorphisms between different coproduct structures and construct a pre-Lie algebra structure on decorated planar rooted forests.

math.CO↗

arXiv2025

Recent Advances in Microfluidics and Bioelectronics for Three-Dimensional Organoid Interfaces

Organoids offer a promising alternative in biomedical research and clinical medicine, with better feature recapitulation than 2D cultures. They also have more consistent responses with clinical results when compared to animal models. However, major challenges exist in the longevity of culture, the reproducibility of organoid properties, and the development of non-disruptive monitoring methods. Recent advances in materials and microfabrication methods, such as 3D printing and compressive buckling, have enabled three-dimensional (3D) interfaces of microfluidics and bioelectronics to manipulate and monitor these biological models in exciting ways. These advanced systems have great potential for applications in drug delivery, personalized medicine, and disease modelling. We conclude with important future considerations to generate longevity using further technological development in organoid and spheroid models.

q-bio.TO↗

arXiv2025

Productions of $T_{cc}$ and its SU(3)-flavor symmetry and heavy quark spin symmetry partners in $B_c$ decays

Inspired by the observation of the doubly charmed tetraquark state $T_{cc}$ at $pp$ collisions in the inclusive processes, we systematically investigate the production of doubly charmed tetraquark states in exclusive $B_c$ decays. In this work, we assume the $T_{cc}$ as a $DD^*$ bound state, and then predict the masses of its heavy quark spin symmetry partner $D^*D^*$ (denoted by $T_{cc}^{*}$) as well as their SU(3)-flavor symmetry partners, i.g., $D_sD^*/D_s^*D$ (denoted by $T_{ccs}^{+}$ and $T_{ccs}^{++}$) and $D^*D_s^*$ (denoted by $T_{ccs}^{*+}$ and $T_{ccs}^{*++}$), using the contact range effective field theory. Within the molecular picture, we compute their partial decays and production rates in $B_c$ meson decays. We identify the decays $B_c \to D^0D^0π^+\bar{D}^0 $ and $B_c \to D_s^+ D^{0} π^+{D}^{-}$ as promising channels to observe the tetraquark states $T_{cc}^{(*)}$ and $T_{ccs}^{(*)++}$, respectively. Finally, by combining these results with $B_c$ production cross sections, we estimate the expected event yields for these states in the upcoming LHC Run 3 and 4. Our results indicate that the $T_{ccs}^{+}$ and $T_{ccs}^{*++}$ states are likely to be observed in $B_c$ decays, while it is quite difficult for the $T_{cc}$ and $T_{cc}^{*}$ states.

hep-ph↗

arXiv2025

CardRewriter: Leveraging Knowledge Cards for Long-Tail Query Rewriting on Short-Video Platforms

Short-video platforms have rapidly become a new generation of information retrieval systems, where users formulate queries to access desired videos. However, user queries, especially long-tail ones, often suffer from spelling errors, incomplete phrasing, and ambiguous intent, resulting in mismatches between user expectations and retrieved results. While large language models (LLMs) have shown success in long-tail query rewriting within e-commerce, they struggle on short-video platforms, where proprietary content such as short videos, live streams, micro dramas, and user social networks falls outside their training distribution. To address this challenge, we introduce \textbf{CardRewriter}, an LLM-based framework that incorporates domain-specific knowledge to enhance long-tail query rewriting. For each query, our method aggregates multi-source knowledge relevant to the query and summarizes it into an informative and query-relevant knowledge card. This card then guides the LLM to better capture user intent and produce more effective query rewrites. We optimize CardRewriter using a two-stage training pipeline: supervised fine-tuning followed by group relative policy optimization, with a tailored reward system balancing query relevance and retrieval effectiveness. Offline experiments show that CardRewriter substantially improves rewriting quality for queries targeting proprietary content. Online A/B testing further confirms significant gains in long-view rate (LVR) and click-through rate (CTR), along with a notable reduction in initiative query reformulation rate (IQRR). Since September 2025, CardRewriter has been deployed on Kuaishou, one of China's largest short-video platforms, serving hundreds of millions of users daily.

cs.IR↗

arXiv2025

RAGDiffusion: Faithful Cloth Generation via External Knowledge Assimilation

Standard clothing asset generation involves restoring forward-facing flat-lay garment images displayed on a clear background by extracting clothing information from diverse real-world contexts, which presents significant challenges due to highly standardized structure sampling distributions and clothing semantic absence in complex scenarios. Existing models have limited spatial perception, often exhibiting structural hallucinations and texture distortion in this high-specification generative task. To address this issue, we propose a novel Retrieval-Augmented Generation (RAG) framework, termed RAGDiffusion, to enhance structure determinacy and mitigate hallucinations by assimilating knowledge from language models and external databases. RAGDiffusion consists of two processes: (1) Retrieval-based structure aggregation, which employs contrastive learning and a Structure Locally Linear Embedding (SLLE) to derive global structure and spatial landmarks, providing both soft and hard guidance to counteract structural ambiguities; and (2) Omni-level faithful garment generation, which introduces a coarse-to-fine texture alignment that ensures fidelity in pattern and detail components within the diffusing. Extensive experiments on challenging real-world datasets demonstrate that RAGDiffusion synthesizes structurally and texture-faithful clothing assets with significant performance improvements, representing a pioneering effort in high-specification faithful generation with RAG to confront intrinsic hallucinations and enhance fidelity.

cs.CV↗

arXiv2025

Solving Fokker-Planck-Kolmogorov Equation by Distribution Self-adaptation Normalized Physics-informed Neural Networks

Stochastic dynamical systems provide essential mathematical frameworks for modeling complex real-world phenomena. The Fokker-Planck-Kolmogorov (FPK) equation governs the evolution of probability density functions associated with stochastic system trajectories. Developing robust numerical methods for solving the FPK equation is critical for understanding and predicting stochastic behavior. Here, we introduce the distribution self-adaptive normalized physics-informed neural network (DSN-PINNs) for solving time-dependent FPK equations through the integration of soft normalization constraints with adaptive resampling strategies. Specifically, we employ a normalization-enhanced PINN model in a pretraining phase to establish the solution's global structure and scale, generating a reliable prior distribution. Subsequently, guided by this prior, we dynamically reallocate training points via weighted kernel density estimation, concentrating computational resources on regions most representative of the underlying probability distribution throughout the learning process. The key innovation lies in our method's ability to exploit the intrinsic structural properties of stochastic dynamics while maintaining computational accuracy and implementation simplicity. We demonstrate the framework's effectiveness through comprehensive numerical experiments and comparative analyses with existing methods, including validation on real-world economic datasets.

stat.CO↗

arXiv2025

The Mottness and the Anderson localization in bilayer nickelate La$_3$Ni$_2$O$_{7-δ}$

The oxygen content plays a pivotal role in determining the electronic and superconducting properties of the recently discovered La$_3$Ni$_2$O$_{7-δ}$ superconductors. In this work, we investigate the impact of oxygen vacancies on the insulating behavior of La$_3$Ni$_2$O$_{7-δ}$ across the doping range $δ= 0$ to $0.5$. At $δ= 0.5$, we construct a bilayer two-orbital Hubbard model to describe the system. Using dynamical mean-field theory, we demonstrate that the model captures the characteristics of a bilayer Mott insulator. To explore the effects of disorder within the range $δ= 0$ to $0.5$, we treat the system as a mixture of metallic and Mott insulating phases. By applying the dynamical cluster approximation and the typical medium dynamical cluster approximation, we identify an Anderson localization transition at a critical doping of $δ\sim 0.2$ through the geometric average of the local density of states. This Anderson localization transition is the key reason for the suppression of superconductivity in La$_3$Ni$_2$O$_{7-δ}$. These results provide a quantitative explanation of recent experimental observations and highlight the critical influence of oxygen content on the physical properties of La$_3$Ni$_2$O$_{7-δ}$.

cond-mat.str-el↗

arXiv2025

IPR: Intelligent Prompt Routing with User-Controlled Quality-Cost Trade-offs

Routing incoming queries to the most cost-effective LLM while maintaining response quality poses a fundamental challenge in optimizing performance-cost trade-offs for large-scale commercial systems. We present IPR\, -- \,a quality-constrained \textbf{I}ntelligent \textbf{P}rompt \textbf{R}outing framework that dynamically selects optimal models based on predicted response quality and user-specified tolerance levels. IPR introduces three key innovations: (1) a modular architecture with lightweight quality estimators trained on 1.5M prompts annotated with calibrated quality scores, enabling fine-grained quality prediction across model families; (2) a user-controlled routing mechanism with tolerance parameter $τ\in [0,1]$ that provides explicit control over quality-cost trade-offs; and (3) an extensible design using frozen encoders with model-specific adapters, reducing new model integration from days to hours. To rigorously train and evaluate IPR, we curate an industrial-level dataset IPRBench\footnote{IPRBench will be released upon legal approval.}, a comprehensive benchmark containing 1.5 million examples with response quality annotations across 11 LLM candidates. Deployed on a major cloud platform, IPR achieves 43.9\% cost reduction while maintaining quality parity with the strongest model in the Claude family and processes requests with sub-150ms latency. The deployed system and additional product details are publicly available at https://aws.amazon.com/bedrock/intelligent-prompt-routing/

cs.LG↗

arXiv2025

Co-design is powerful and not free

Robotic performance emerges from the coupling of body and controller, yet it remains unclear when morphology-control co-design is necessary. We present a unified framework that embeds morphology and control parameters within a single neural network, enabling end-to-end joint optimization. Through case studies in static-obstacle-constrained reaching, we evaluate trajectory error, success rate, and collision probability. The results show that co-design provides clear benefits when morphology is poorly matched to the task, such as near obstacles or workspace boundaries, where structural adaptation simplifies control. Conversely, when the baseline morphology already affords sufficient capability, control-only optimization often matches or exceeds co-design. By clarifying when control is enough and when it is not, this work advances the understanding of embodied intelligence and offers practical guidance for embodiment-aware robot design.

cs.NE↗

arXiv2025

Improving Retrieval-Augmented Generation through Multi-Agent Reinforcement Learning

Retrieval-augmented generation (RAG) is widely utilized to incorporate external knowledge into large language models, thereby enhancing factuality and reducing hallucinations in question-answering (QA) tasks. A standard RAG pipeline consists of several components, such as query rewriting, document retrieval, document filtering, and answer generation. However, these components are typically optimized separately through supervised fine-tuning, which can lead to misalignments between the objectives of individual components and the overarching aim of generating accurate answers. Although recent efforts have explored using reinforcement learning (RL) to optimize specific RAG components, these approaches often focus on simple pipelines with only two components or do not adequately address the complex interdependencies and collaborative interactions among the modules. To overcome these limitations, we propose treating the complex RAG pipeline with multiple components as a multi-agent cooperative task, in which each component can be regarded as an RL agent. Specifically, we present MMOA-RAG, Multi-Module joint Optimization Algorithm for RAG, which employs multi-agent reinforcement learning to harmonize all agents' goals toward a unified reward, such as the F1 score of the final answer. Experiments conducted on various QA benchmarks demonstrate that MMOA-RAG effectively boost the overall performance of the pipeline and outperforms existing baselines. Furthermore, comprehensive ablation studies validate the contributions of individual components and demonstrate MMOA-RAG can be adapted to different RAG pipelines and benchmarks.

cs.CL↗

arXiv2025

The hot circumgalactic medium in stacked X-rays: observations vs simulations

Current cosmological simulations rely on active galactic nuclei (AGN) feedback to quench star formation and match observed stellar mass distributions, but models for AGN feedback are poorly constrained. The circumgalactic medium (CGM) provides a valuable laboratory to study this process, as its metallicity, temperature, and density distributions are directly shaped by AGN activity. Recent observations from the eROSITA instrument provide constraints on the CGM through measurements of extended soft X-ray emission. In this work, we generate synthetic eROSITA observations from the EAGLE and SIMBA cosmological simulations and compare them to observations of galaxies stacked by stellar mass, halo mass, and star-formation rate. SIMBA outperforms EAGLE in matching observed surface brightness profiles, but neither simulation achieves consistent agreement with observations across the full range of galaxy properties we studied. We find that variations in CGM X-ray emission between simulations are primarily driven by density differences at $R \lesssim 0.2 R_{200c} $, and temperature and metallicity changes at larger radii. These results highlight the need for further refinement of AGN feedback models in cosmological simulations and demonstrate the power of stacked X-ray observations as a tool for constraining feedback physics.

astro-ph.GA↗

arXiv2025

MARCO: A Cooperative Knowledge Transfer Framework for Personalized Cross-domain Recommendations

Recommender systems frequently encounter data sparsity issues, particularly when addressing cold-start scenarios involving new users or items. Multi-source cross-domain recommendation (CDR) addresses these challenges by transferring valuable knowledge from multiple source domains to enhance recommendations in a target domain. However, existing reinforcement learning (RL)-based CDR methods typically rely on a single-agent framework, leading to negative transfer issues caused by inconsistent domain contributions and inherent distributional discrepancies among source domains. To overcome these limitations, MARCO, a Multi-Agent Reinforcement Learning-based Cross-Domain recommendation framework, is proposed. It leverages cooperative multi-agent reinforcement learning, where each agent is dedicated to estimating the contribution from an individual source domain, effectively managing credit assignment and mitigating negative transfer. In addition, an entropy-based action diversity penalty is introduced to enhance policy expressiveness and stabilize training by encouraging diverse agents' joint actions. Extensive experiments across four benchmark datasets demonstrate MARCO's superior performance over state-of-the-art methods, highlighting its robustness and strong generalization capabilities. The code is at https://github.com/xiewilliams/MARCO.

cs.IR↗

arXiv2025

$B_{(s)} \to S(a_0(1450), K_0^*(1430), f_0(1500))$ helicity form factors within QCD light-cone sum rules

In this paper, we investigate the helicity form factors (HFFs) of the $B_{(s)}$-meson decay into a scalar meson with a mass larger than 1~GeV, {\it i.e.,} $B \to a_0(1450)$, $B_{(s)} \to K_0^*(1430)$ and $B_{s} \to f_0(1500)$ by using light-cone sum rules approach. We take the standard currents for correlation functions. To enhance the precision of our calculations, we incorporate the next-to-leading order (NLO) corrections and retain the scalar meson twist-3 light-cone distribution amplitudes. Furthermore, we extend the HFFs to the entire physical $q^2$ region employing a simplified $z$-series expansion. At the point of $q^2=1\rm{~GeV^2}$, all NLO contributions to the HFFs are negative, with the maximum contribution around $25\%$. Then, as applications of these HFFs, we analyze the differential decay widths, branching ratios, and lepton polarization asymmetries for the semi-leptonic $B_{(s)} \to S \ell \barν_\ell$, FCNC $B_{(s)} \to S \ell \bar{\ell}$ and rare $B_{(s)} \to S ν\barν$ decays. Our results are consistent with existing studies within uncertainties. The current data still suffer from large uncertainties and need to be measured more precisely, which can lead to a better understanding of the fundamental properties of light scalar mesons.

hep-ph↗