arXiv subjects

Yu Zhang

Publications and source records attributed to Yu Zhang.

At least 451 records · Page 25Linked to original sources

arXiv2025

StarWhisper Telescope: An AI framework for automating end-to-end astronomical observations

The exponential growth of large-scale telescope arrays has boosted time-domain astronomy development but introduced operational bottlenecks, including labor-intensive observation planning, data processing, and real-time decision-making. Here we present the StarWhisper Telescope system, an AI agent framework automating end-to-end astronomical observations for surveys like the Nearby Galaxy Supernovae Survey. By integrating large language models with specialized function calls and modular workflows, StarWhisper Telescope autonomously generates site-specific observation lists, executes real-time image analysis via pipelines, and dynamically triggers follow-up proposals upon transient detection. The system reduces human intervention through automated observation planning, telescope controlling and data processing, while enabling seamless collaboration between amateur and professional astronomers. Deployed across Nearby Galaxy Supernovae Survey's network of 10 amateur telescopes, the StarWhisper Telescope has detected transients with promising response times relative to existing surveys. Furthermore, StarWhisper Telescope's scalable agent architecture provides a blueprint for future facilities like the Global Open Transient Telescope Array, where AI-driven autonomy will be critical for managing 60 telescopes.

astro-ph.IM↗

arXiv2025

Temporal-order-driven asymmetric quantum interference and temporal coherence enhancement in spontaneous six-wave mixing

Narrow-band multiphoton entanglement sources serve as a core enabling resource for advanced quantum information technologies. Recently, researchers have directly generated energy-time entangled triphoton W states in a hot atomic medium via spontaneous six-wave mixing for the first time. However, a rigorous theoretical framework for this process remains lacking to date, confining our understanding to a mere extension of the biphoton model. Here, we analytically investigate the generation mechanism of energy-time entangled triphotons and their classically controllable optical properties in an electromagnetically induced transparency-assisted five-level cold atomic system. Notably, triphoton generation follows strict temporal ordering, resulting in asymmetric quantum interference in triple coincidence counts--unreplicable and unexplainable by the inherently symmetric biphoton model. These results establish a rigorous physical framework for spontaneous six-wave mixing-generated triphotons, clarify their distinctions from states produced via cascaded nonlinear models, and substantially advance their utility in quantum information protocols.

quant-ph↗

arXiv2025

Copy-Augmented Representation for Structure Invariant Template-Free Retrosynthesis

Retrosynthesis prediction is fundamental to drug discovery and chemical synthesis, requiring the identification of reactants that can produce a target molecule. Current template-free methods struggle to capture the structural invariance inherent in chemical reactions, where substantial molecular scaffolds remain unchanged, leading to unnecessarily large search spaces and reduced prediction accuracy. We introduce C-SMILES, a novel molecular representation that decomposes traditional SMILES into element-token pairs with five special tokens, effectively minimizing editing distance between reactants and products. Building upon this representation, we incorporate a copy-augmented mechanism that dynamically determines whether to generate new tokens or preserve unchanged molecular fragments from the product. Our approach integrates SMILES alignment guidance to enhance attention consistency with ground-truth atom mappings, enabling more chemically coherent predictions. Comprehensive evaluation on USPTO-50K and large-scale USPTO-FULL datasets demonstrates significant improvements: 67.2% top-1 accuracy on USPTO-50K and 50.8% on USPTO-FULL, with 99.9% validity in generated molecules. This work establishes a new paradigm for structure-aware molecular generation with direct applications in computational drug discovery.

cs.LG↗

arXiv2025

MRSAudio: A Large-Scale Multimodal Recorded Spatial Audio Dataset with Refined Annotations

Humans rely on multisensory integration to perceive spatial environments, where auditory cues enable sound source localization in three-dimensional space. Despite the critical role of spatial audio in immersive technologies such as VR/AR, most existing multimodal datasets provide only monaural audio, which limits the development of spatial audio generation and understanding. To address these challenges, we introduce MRSAudio, a large-scale multimodal spatial audio dataset designed to advance research in spatial audio understanding and generation. MRSAudio spans four distinct components: MRSLife, MRSSpeech, MRSMusic, and MRSSing, covering diverse real-world scenarios. The dataset includes synchronized binaural and ambisonic audio, exocentric and egocentric video, motion trajectories, and fine-grained annotations such as transcripts, phoneme boundaries, lyrics, scores, and prompts. To demonstrate the utility and versatility of MRSAudio, we establish five foundational tasks: audio spatialization, and spatial text to speech, spatial singing voice synthesis, spatial music generation and sound event localization and detection. Results show that MRSAudio enables high-quality spatial modeling and supports a broad range of spatial audio research. Demos and dataset access are available at https://mrsaudio.github.io.

cs.SD↗

arXiv2025

DCMIL: A Progressive Representation Learning of Whole Slide Images for Cancer Prognosis Analysis

The burgeoning discipline of computational pathology shows promise in harnessing whole slide images (WSIs) to quantify morphological heterogeneity and develop objective prognostic modes for human cancers. However, progress is impeded by the computational bottleneck of gigapixel-size inputs and the scarcity of dense manual annotations. Current methods often overlook fine-grained information across multi-magnification WSIs and variations in tumor microenvironments. Here, we propose an easy-to-hard progressive representation learning, termed dual-curriculum contrastive multi-instance learning (DCMIL), to efficiently process WSIs for cancer prognosis. The model does not rely on dense annotations and enables the direct transformation of gigapixel-size WSIs into outcome predictions. Extensive experiments on twelve cancer types (5,954 patients, 12.54 million tiles) demonstrate that DCMIL outperforms standard WSI-based prognostic models. Additionally, DCMIL identifies fine-grained prognosis-salient regions, provides robust instance uncertainty estimation, and captures morphological differences between normal and tumor tissues, with the potential to generate new biological insights. All codes have been made publicly accessible at https://github.com/tuuuc/DCMIL.

cs.CV↗

arXiv2025

Symmetry-forbidden intraband transitions leading to ultralow Gilbert damping in van der Waals ferromagnets

Based upon first-principles calculations, we report ultralow Gilbert damping in two-dimensional (2D) van derWaals (vdW) ferromagnets. The low damping occurs at weak scattering because mirror symmetry prohibits intraband transitions. The monotonic dependence on the electronic scattering rate suggests the absent lower limit, in contrast to conventional ferromagnetic materials. Breaking mirror symmetry through magnetization rotation, layer stacking, or structural phase transition significantly increases damping by enabling intraband transitions. Topological nodal lines, also protected by mirror symmetry, contribute substantially to interband-transition-mediated damping, which can be tuned by adjusting the Fermi level. Our findings elucidate the unique characteristics of Gilbert damping in 2D vdW ferromagnets, providing valuable insights for designing low-dimensional spintronic devices with high energy efficiency.

cond-mat.mtrl-sci↗

arXiv2025

Synergistic Integration and Discrepancy Resolution of Contextualized Knowledge for Personalized Recommendation

The integration of large language models (LLMs) into recommendation systems has revealed promising potential through their capacity to extract world knowledge for enhanced reasoning capabilities. However, current methodologies that adopt static schema-based prompting mechanisms encounter significant limitations: (1) they employ universal template structures that neglect the multi-faceted nature of user preference diversity; (2) they implement superficial alignment between semantic knowledge representations and behavioral feature spaces without achieving comprehensive latent space integration. To address these challenges, we introduce CoCo, an end-to-end framework that dynamically constructs user-specific contextual knowledge embeddings through a dual-mechanism approach. Our method realizes profound integration of semantic and behavioral latent dimensions via adaptive knowledge fusion and contradiction resolution modules. Experimental evaluations across diverse benchmark datasets and an enterprise-level e-commerce platform demonstrate CoCo's superiority, achieving a maximum 8.58% improvement over seven cutting-edge methods in recommendation accuracy. The framework's deployment on a production advertising system resulted in a 1.91% sales growth, validating its practical effectiveness. With its modular design and model-agnostic architecture, CoCo provides a versatile solution for next-generation recommendation systems requiring both knowledge-enhanced reasoning and personalized adaptation.

cs.IR↗

arXiv2025

FedLoRA-Optimizer: Federated LoRA Fine-Tuning with Global and Local Optimization in Heterogeneous Data Scenarios

Federated efficient fine-tuning has emerged as an approach that leverages distributed data and computational resources across nodes to address the challenges of large-scale fine-tuning and privacy preservation. The Low-Rank Adaptation (LoRA) enables efficient fine-tuning of large-scale pre-trained models by introducing trainable low-rank matrices into weight updates.However, in heterogeneous data scenarios, client drift weakens the generalization of the global model, and local models often fail to meet the personalized needs of individual clients.Moreover, existing federated LoRA efficient fine-tuning techniques overlook fine-grained analysis of the tuning matrices. To address this, we conducted preliminary experiments and found that different LoRA matrices exhibit different sensitivity to changes in the direction and magnitude of their vectors.We thus propose a fine-grained federated LoRA tuning method. By fine-tuning the more sensitive directional vectors in the A matrix, which encode shared knowledge, our method learns shared features more effectively across clients and enhances global generalization. Simultaneously, by fine-tuning the more sensitive magnitude vectors in the B matrix, which encode personalized knowledge, our method better captures personalized knowledge, enabling detailed adaptation to local data. The method uses a pipeline combining global and local optimizers. Global optimization further improves local models, achieving collaborative optimization between global and local levels. This improves both the generalization ability of the global model and the personalized adaptation of local models under heterogeneous data scenarios. Experiments on Databricks-Dolly-15k and Natural Instructions with LLaMA2-7B and Deepseek-7B confirm that our method improves global performance by 0.39% and local performance by 0.59%.

cs.LG↗

arXiv2025

Student-AI Interaction in an LLM-Empowered Learning Environment: A Cluster Analysis of Engagement Profiles

Integrating Large Language Models (LLMs) into educational practice enables personalized learning by accommodating diverse learner behaviors. This study explored diverse learner profiles within a multi-agent, LLM-empowered learning environment. Data was collected from 312 undergraduate students at a university in China as they participated in a six-module course. Based on hierarchical cluster analyses of system profiles and student-AI interactive dialogues, we found that students exhibit varied behavioral, cognitive, and emotional engagement tendencies. This analysis allowed us to identify two types of dropouts (early dropouts and stagnating interactors) and three completer profiles (active questioners, responsive navigators, and lurkers). The results showed that high levels of interaction do not always equate to productive learning and vice versa. Prior knowledge significantly influenced interaction patterns and short-term learning benefits. Further analysis of the human-AI dialogues revealed that some students actively engaged in knowledge construction, while others displayed a high frequency of regulatory behaviors. Notably, both groups of students achieved comparable learning gains, demonstrating the effectiveness of the multi-agent learning environment in supporting personalized learning. These results underscore the complex and multifaceted nature of engagement in human-AI collaborative learning and provide practical implications for the design of adaptive educational systems.

cs.CY↗

arXiv2025

SpatialSplat: Efficient Semantic 3D from Sparse Unposed Images

A major breakthrough in 3D reconstruction is the feedforward paradigm to generate pixel-wise 3D points or Gaussian primitives from sparse, unposed images. To further incorporate semantics while avoiding the significant memory and storage costs of high-dimensional semantic features, existing methods extend this paradigm by associating each primitive with a compressed semantic feature vector. However, these methods have two major limitations: (a) the naively compressed feature compromises expressiveness, affecting the model's ability to capture fine-grained semantics, and (b) the pixel-wise primitive prediction introduces redundancy in overlapping areas, causing unnecessary memory overhead. To this end, we introduce \textbf{SpatialSplat}, a feedforward framework that produces redundancy-aware Gaussians and capitalizes on a dual-field semantic representation. Particularly, with the insight that primitives within the same instance exhibit high semantic consistency, we decompose the semantic representation into a coarse feature field that encodes uncompressed semantics with minimal primitives, and a fine-grained yet low-dimensional feature field that captures detailed inter-instance relationships. Moreover, we propose a selective Gaussian mechanism, which retains only essential Gaussians in the scene, effectively eliminating redundant primitives. Our proposed Spatialsplat learns accurate semantic information and detailed instances prior with more compact 3D Gaussians, making semantic 3D reconstruction more applicable. We conduct extensive experiments to evaluate our method, demonstrating a remarkable 60\% reduction in scene representation parameters while achieving superior performance over state-of-the-art methods. The code is available at https://github.com/shengyuuu/SpatialSplat.git

cs.CV↗

arXiv2025

Bridging Control Variates and Regression Adjustment in A/B Testing: From Design-Based to Model-Based Frameworks

A B testing serves as the gold standard for large scale, data driven decision making in online businesses. To mitigate metric variability and enhance testing sensitivity, control variates and regression adjustment have emerged as prominent variance reduction techniques, leveraging pre experiment data to improve estimator performance. Over the past decade, these methods have spawned numerous derivatives, yet their theoretical connections and comparative properties remain underexplored. In this paper, we conduct a comprehensive analysis of their statistical properties, establish a formal bridge between the two frameworks in practical implementations, and extend the investigation from design based to model-based frameworks. Through simulation studies and real world experiments at ByteDance, we validate our theoretical insights across both frameworks. Our work aims to provide rigorous guidance for practitioners in online controlled experiments, addressing critical considerations of internal and external validity. The recommended method control variates with group specific coefficient estimates has been fully implemented and deployed on ByteDance's experimental platform.

stat.ME↗

arXiv2025

Student Development Agent: Risk-free Simulation for Evaluating AIED Innovations

In the age of AI-powered educational (AIED) innovation, evaluating the developmental consequences of novel designs before they are exposed to students has become both essential and challenging. Since such interventions may carry irreversible effects, it is critical to anticipate not only potential benefits but also possible harms. This study proposes a student development agent framework based on large language models (LLMs), designed to simulate how students with diverse characteristics may evolve under different educational settings without administering them to real students. By validating the approach through a case study on a multi-agent learning environment (MAIC), we demonstrate that the agent's predictions align with real student outcomes in non-cognitive developments. The results suggest that LLM-based simulations hold promise for evaluating AIED innovations efficiently and ethically. Future directions include enhancing profile structures, incorporating fine-tuned or small task-specific models, validating effects of empirical findings, interpreting simulated data and optimizing evaluation methods.

cs.CY↗

arXiv2025

Observation of multiple time crystals in a driven-dissipative system with Rydberg gas

Time crystals, as temporal analogs of space crystals, manifest as stable and periodic behavior that breaks time translation symmetry. In an open quantum system, many-body interaction subjected to dissipation allows one to develop the time crystalline order in an unprecedented way, as refer to dissipative time crystals. Here we report the observation of multiple time crystals in the continuously driven-dissipative and strongly interacting Rydberg thermal gases, in which continuous time crystals, sub-harmonic time crystals, and high-harmonic time crystals are observed in the same system by manipulating the Rydberg excitation. Our work provides new ways to explore the nonequilibrium phases of matter in open systems. Such time crystals with persistent oscillation rooted in emergent quantum correlations, may emerge as a ubiquitous tool in quantum metrology, for instance, continuous sensing and parameter estimation surpassing the standard quantum limit.

physics.atom-ph↗

arXiv2025

SRPO: Enhancing Multimodal LLM Reasoning via Reflection-Aware Reinforcement Learning

Multimodal large language models (MLLMs) have shown promising capabilities in reasoning tasks, yet still struggle with complex problems requiring explicit self-reflection and self-correction, especially compared to their unimodal text-based counterparts. Existing reflection methods are simplistic and struggle to generate meaningful and instructive feedback, as the reasoning ability and knowledge limits of pre-trained models are largely fixed during initial training. To overcome these challenges, we propose Multimodal Self-Reflection enhanced reasoning with Group Relative Policy Optimization (SRPO), a two-stage reflection-aware reinforcement learning (RL) framework explicitly designed to enhance multimodal LLM reasoning. In the first stage, we construct a high-quality, reflection-focused dataset under the guidance of an advanced MLLM, which generates reflections based on initial responses to help the policy model learn both reasoning and self-reflection. In the second stage, we introduce a novel reward mechanism within the GRPO framework that encourages concise and cognitively meaningful reflection while avoiding redundancy. Extensive experiments across multiple multimodal reasoning benchmarks, including MathVista, MathVision, MathVerse, and MMMU-Pro, using Qwen-2.5-VL-7B and Qwen-2.5-VL-32B demonstrate that SRPO significantly outperforms state-of-the-art models, achieving notable improvements in both reasoning accuracy and reflection quality.

cs.CL↗

arXiv2025

ElasticMoE: An Efficient Auto Scaling Method for Mixture-of-Experts Models

Mixture-of-Experts (MoE) models promise efficient scaling of large language models (LLMs) by activating only a small subset of experts per token, but their parallelized inference pipelines make elastic serving challenging. Existing strategies fall short: horizontal scaling provisions entire replicas of the current configuration, often tens to hundreds of accelerators, leading to coarse granularity, long provisioning delays, and costly overprovisioning. Vertical scaling offers finer adjustments but typically requires instance restarts, incurring downtime. These limitations make current approaches ill-suited for the bursty, short-lived traffic patterns common in cloud deployments. We present ElasticMoE, an elastic scaling framework for MoE LLMs that achieves fine-grained, low-latency, and zero-downtime scaling. ElasticMoE decouples inference execution from memory operations, enabling scaling steps to proceed concurrently with serving. An HBM Management Module (HMM) reuses weights and KV caches via zero-copy remapping, while high-bandwidth peer-to-peer transfers bring newly added accelerators online without interrupting service. A virtual memory based expert redistribution mechanism migrates MoE experts without costly buffer reallocations, reducing peak memory usage during expert parallelism reconfiguration. Our evaluation on Ascend NPUs with three popular MoE LLMs shows that ElasticMoE achieves up to 9x lower scale-up latency, up to 2x better throughput during scaling, and significantly improves SLO attainment compared to baselines. By enabling fine-grained, concurrent scaling with minimal disruption, ElasticMoE advances the practicality of deploying massive MoE LLMs in dynamic cloud environments.

cs.DC↗

arXiv2025

Alternating Training-based Label Smoothing Enhances Prompt Generalization

Recent advances in pre-trained vision-language models have demonstrated remarkable zero-shot generalization capabilities. To further enhance these models' adaptability to various downstream tasks, prompt tuning has emerged as a parameter-efficient fine-tuning method. However, despite its efficiency, the generalization ability of prompt remains limited. In contrast, label smoothing (LS) has been widely recognized as an effective regularization technique that prevents models from becoming over-confident and improves their generalization. This inspires us to explore the integration of LS with prompt tuning. However, we have observed that the vanilla LS even weakens the generalization ability of prompt tuning. To address this issue, we propose the Alternating Training-based Label Smoothing (ATLaS) method, which alternately trains with standard one-hot labels and soft labels generated by LS to supervise the prompt tuning. Moreover, we introduce two types of efficient offline soft labels, including Class-wise Soft Labels (CSL) and Instance-wise Soft Labels (ISL), to provide inter-class or instance-class relationships for prompt tuning. The theoretical properties of the proposed ATLaS method are analyzed. Extensive experiments demonstrate that the proposed ATLaS method, combined with CSL and ISL, consistently enhances the generalization performance of prompt tuning. Moreover, the proposed ATLaS method exhibits high compatibility with prevalent prompt tuning methods, enabling seamless integration into existing methods.

cs.CV↗

arXiv2025

Strongly Entangled Kondo and Kagome Lattices and the Emergent Magnetic Ground State in Heavy-Fermion Kagome Metal YbV$_6$Sn$_6$

Applying angle-resolved photoemission spectroscopy and density functional theory calculations, we present compelling spectroscopic evidence demonstrating the intertwining and mutual interaction between the Kondo and kagome sublattices in heavy-fermion intermetallic compound YbV$_6$Sn$_6$. We reveal the Yb 4$f$-derived states near the Fermi level, along with the presence of bulk kagome bands and topological surface states. We unveil strong interactions between the 4$f$ and itinerant electrons, where the kagome bands hosting the Dirac fermions and van Hove singularities predominate. Such findings are well described using a $c$-$f$ hybridization model. On the other hand, our systematic characterization of magnetic properties demonstrates an unusually enhanced antiferromagnetic ordering, where the kagome-derived van Hove singularities near $E_F$ play a vital role in determining the unconventional nature of the Ruderman-Kittel-Kasuya-Yosida interaction and Kondo coupling. These unique kagome-state-mediated exchange interactions have never been reported before and could lead to a novel phase diagram and various quantum critical behaviors in YbV$_6$Sn$_6$ and its siblings. Our results not only expand the family of exotic quantum phases entangled with kagome structure to the strongly correlated regime, but also establish YbV$_6$Sn$_6$ as an unprecedented platform to explore unconventional many-body physics beyond the standard Kondo picture.

cond-mat.str-el↗

arXiv2025

Active Shadowing (ASD): Manipulating Perception of Robotic Behaviors via Implicit Virtual Communication

Explicit communication is often valued for its directness in presenting information but requires attention during exchange, resulting in cognitive interruptions. On the other hand, implicit communication contributes to tacit and smooth interaction, making it more suitable for teaming, but requires inference for interpretation. This paper studies a novel type of implicit visual communication (IVC) using shadows via visual projection with augmented reality, referred to as active shadowing (ASD). Prior IVC methods, such as legible motion, are often used to influence the perception of robot behavior to make it more understandable. They often require changing the physical robot behavior, resulting in suboptimality. In our work, we investigate how ASD can be used to achieve similar effects without losing optimality. Our evaluations with user studies demonstrates that ASD can effectively creates ''illusions'' that maintain optimal physical behavior without compromising its understandability. We also show that ASD can be more informative than other explicit communication methods, and examine the conditions under which ASD becomes less effective.

cs.RO↗