arXiv subjects

Yuanyuan Xu

Publications and source records attributed to Yuanyuan Xu.

At least 19 recordsLinked to original sources

arXiv2026

Raman Scattering Evidence for Fluctuating Kagome-Plane Moments in $\mathrm{B} \parallel [111]$ in Pr$_2$Zr$_2$O$_7$ pyrochlore

The exotic magnetic properties of the pyrochlore Pr$_2$Zr$_2$O$_7$ are determined by the coupling of spin, orbital, and lattice degrees of freedom. Magnetism in this material originates from the non-Kramers Pr$^{3+}$ ion, and the exotic behavior has been discussed both in the framework of dipole-quadrupole properties of magnetic moments and disorder splitting the non-Kramers doublet. We use magneto-Raman spectroscopy to probe the crystal electric field (CEF) excitations of Pr$^{3+}$ in magnetic fields up to 14 T applied along the [100] and [111] crystallographic directions at 2 K. For $\mathrm{B}\parallel[100]$, the field evolution of the Raman active crystal field modes is quantitatively described by conventional Zeeman splitting of the Pr$^{3+}$ ground state doublet, including the thermally populated upper Zeeman branch at elevated temperature. For $\mathrm{}{B}\parallel[111]$, the spectra separate into responses from the triangular and kagome sublattices, and the CEF excitations of Pr$^{3+}$ on the triangular lattice exhibit the expected Zeeman shift consistent with full moments polarization. In contrast, the CEF excitations of Pr$^{3+}$ on the kagome lattice do not split as expected and instead broaden strongly with increasing field. This behavior is inconsistent with a static polarized configuration and is captured phenomenologically by a motional-narrowing description in which the kagome plane moments fluctuate between Zeeman-split levels on a meV timescale. This result offers a natural explanation for the reduced magnetization observed in $\mathrm{}{B}\parallel[111]$ and provides further evidence of the dipole-quadrupole properties of magnetic moments. It highlights Raman scattering as a sensitive probe of exotic spin-orbital dynamics in frustrated magnets.

cond-mat.str-el↗

arXiv2026

AoiZora: Topology-Aware Auto-Parallel Optimization for Inference of Diffusion Transformers

Video diffusion has quickly grown into a key generative serving workload, yet producing each clip demands many denoising iterations over large spatio-temporal latents, which puts low-latency inference out of reach on a single device. A denoising step is therefore typically distributed across multiple accelerators, and TPU sub-slices have become an attractive and practical fabric for doing so. Current auto-parallel systems, however, search almost exclusively over logical device meshes and disregard how a chosen sharding is actually laid out on the physical TPU interconnect -- an oversight that leaves large, topology-dependent performance on the table. We address this gap with AoiZora, a compiler-mediated topology planner built for low-latency video diffusion inference on TPU sub-slices. Its guiding principle is to reconnect logical sharding with physical placement by drawing on different points in the compilation flow: AoiZora first eliminates weak sharding candidates from inexpensive pre-compilation IRs, then compiles only the ones that survive and orders their physical placements using compiled HLO together with a topology-aware communication model. The winning plan is realized along the ordinary compiler path, leaving model code, compiler lowering, collective kernels, and network routing entirely intact. On TPU v5e sub-slices, AoiZora reduces Wan 2.1 one-step denoising latency by as much as 1.42x relative to existing solutions.

cs.DC↗

arXiv2026

Self-Evolving Agents as Dynamic Graph Transformation: A Survey and New Perspective

Large language model (LLM)-based agents are increasingly becoming self-evolving systems that persist across interactions, maintain memories, use tools, acquire skills, refine workflows, and coordinate with other agents. These capabilities make agent states structural and dynamic: entities, relations, attributes, dependencies, and execution structures change with new evidence, feedback, and environmental conditions. Existing graph-agent surveys typically treat graphs as support structures for agent functions rather than as evolving substrates, while self-evolving-agent surveys focus on agent-level mechanisms and rarely discuss graph topology evolution. Thus, the coupling between evolving agent state and dynamic graph topology remains underexplored. This survey connects these two research lines by framing \textit{agent evolution as dynamic graph transformation}. We model agent state as a dynamic graph, where memories, tools, skills, workflows, and inter-agent relations are represented as typed nodes, edges, and subgraphs updated through schema-constrained rewrites. Based on this formulation, we organize existing dynamic-graph-based methods for self-evolving agents into four taxonomies: node/feature evolution, edge/topology evolution, subgraph activation, and cross-component co-evolution. Building on this taxonomy, we propose dynamic graph learning as reusable infrastructure for self-evolving agents and map nine dynamic-graph-learning subfields to agent-evolution capabilities, discussing their adaptations and possible failure modes. Finally, we discuss five types of graph-aware evaluation and governance protocols from a dynamic-graph perspective, which complement end-task evaluation. The goal is to provide a compact structural lens for designing and governing self-evolving agents.

cs.AI↗

arXiv2026

LT2: Linear-Time Looped Transformers

Looped Transformers (LT) have emerged as a powerful architecture by iterating their layers multiple times before decoding the final token. However, pairing them with full attention retains quadratic complexity, making them computationally expensive and slow. We introduce LT2 (Linear-Time Looped Transformers), a family of looped architectures that replace quadratic softmax attention with subquadratic, linear-time attention. We study two variants: LT2-linear with linear attention and LT2-sparse with sparse attention. We find that looping uniquely synergizes with these variants: it enables iterative memory refinement in linear attention and progressively expands the effective receptive field in sparse attention. We formalize these benefits theoretically and demonstrate consistent empirical gains across controlled recall, state-tracking, and language modeling tasks. We then explore LT2-hybrid, which combines different attention variants in a looped setting. Two variants are especially promising: LT2-hybrid (GDN+DSA), which interleaves linear and sparse attention to maximize efficiency and matches the standard looped transformer's quality at fully linear-time cost; and LT2-hybrid (Full+GDN), which interleaves GDN with a small fraction of full attention layers to maximize quality, surpassing the standard looped transformer in both performance and efficiency. We also show how to convert a pre-trained LT into an LT2-hybrid model. With about 1B tokens of training, our converted model, Ouro-hybrid-1.4B, outperforms industry-level 1B models and is competitive with industry-level 4B models while retaining the speed benefits of linear-time attention. Together, these results show a clear path toward making looped transformers more scalable and advancing efficient, capable small language models.

cs.LG↗

arXiv2026

TRACE: Tourism Recommendation with Accountable Citation Evidence

Tourism is a high-stakes setting for conversational recommender systems (CRS): a plausible-sounding suggestion can waste real money and trip time once a traveler acts on it. Existing CRS benchmarks primarily evaluate systems with a single Recall@k score over entity mentions, and tourism-specific resources add spatial or knowledge-graph context, yet none of them couple multi-turn recommendation with verbatim review-span evidence and rejection recovery. This leaves an evaluation gap for tourism recommendation that is simultaneously trustworthy, verifiable, and adaptive: recommend the right point of interest (POI) for multi-aspect preferences (such as cuisine, price, atmosphere, walking distance), justify each suggestion with verifiable evidence from prior visitors so the traveler can act without trial and error, and recover when the first recommendation is rejected mid-dialogue. We introduce TRACE, where each item is a multi-turn tourism recommendation dialogue with review-span citations and explicit rejection turns: 10,000 dialogues over 2,400 Yelp POIs and 34,208 reviews across eight U.S. cities, paired with 14 retrieval, planning, and LLM baselines, along with 25 metrics organized under Accuracy, Grounding, and Recovery. Across these baselines, TRACE reveals the Three-Competency Gap: LLM Zero-Shot leads in closed-set Recall@1 and rejection recovery but cites less densely than retrievers; non-LLM retrievers achieve surface-verbatim grounding but with low accuracy; Multi-Review Synthesis fails at recovery. The Grounding Score agrees with human citation precision (Spearman rho=+0.80, p<10^-20), and paired t-tests reproduce the per-baseline ranking (p<0.01 on the dominant contrasts). TRACE reframes accountable tourism recommendation as a joint target (right POI, verifiable evidence, adaptive repair) rather than a single-axis leaderboard.

cs.IR↗

arXiv2026

HiDE: Hierarchical Dictionary-Based Entropy Modeling for Learned Image Compression

Learned image compression (LIC) has achieved remarkable coding efficiency, where entropy modeling plays a pivotal role in minimizing bitrate through informative priors. Existing methods predominantly exploit internal contexts within the input image, yet the rich external priors embedded in large-scale training data remain largely underutilized. Recent advances in dictionary-based entropy models have demonstrated that incorporating external priors can substantially enhance compression performance. However, current approaches organize heterogeneous external priors within a single-level dictionary, resulting in imbalanced utilization and limited representational capacity. Moreover, effective entropy modeling requires not only expressive priors but also a parameter estimation network capable of interpreting them. To address these challenges, we propose HiDE, a Hierarchical Dictionary-based Entropy modeling framework for learned image compression. HiDE decomposes external priors into global structural and local detail dictionaries with cascaded retrieval, enabling structured and efficient utilization of external information. Moreover, a context-aware parameter estimator with parallel multi-receptive-field design is introduced to adaptively exploit heterogeneous contexts for accurate conditional probability estimation. Experimental results show that HiDE achieves 18.5%, 21.99%, and 24.01% BD-rate savings over VTM-12.1 on the Kodak, CLIC, and Tecnick datasets, respectively.

eess.IV↗

arXiv2026

Mapping Ecological Empathy: A Semantic Network Analysis of Player Perceptions in 3D Environmental Education Games

As the global climate crisis intensifies, 3D video games have emerged as powerful, interactive simulations for Environmental Education (EE). However, empirical assessment of their pedagogical efficacy remains epistemologically challenged. Traditional evaluation metrics, such as pre-post surveys, often suffer from response bias and fail to capture the nuanced, emergent psychological shifts players experience during gameplay. This paper proposes a novel, non-intrusive approach: utilizing Semantic Network Analysis (SNA) to map the 'unsupervised' cognitive structures of players. We scraped and qualitatively filtered 1,825 rich-text user reviews from Steam for two distinct titles representing opposing ecological philosophies: Eco (anthropocentric systemic management) and WolfQuest (biocentric embodied survival). By constructing co-occurrence networks and calculating topological metrics, we visualized the divergence in how players conceptualize human-nature relationships. Results indicate a fundamental pedagogical split: Eco promotes 'Socio-Political Cognition,' where environmental challenges are framed as legislative and economic frictions; conversely, WolfQuest fosters 'Effective Empathy,' where players internalize the fragility of life through the vulnerability of the avatar. We argue that semantic topology offers a rigorous methodological tool for serious games assessment, revealing that effective environmental education requires a strategic tension between systemic logic and emotional resonance.

cs.HC↗

arXiv2026

Struggle as Flow: Challenge, Design, and Experience in Soulslike Games

While traditional game design prioritizes friction-free accessibility, the Soulslike subgenre has achieved commercial dominance through punishing difficulty and frequent failure. This paper challenges the conventional hedonistic paradigm of gaming to investigate the psychological mechanisms behind the Paradox of Failure. By integrating Csikszentmihalyi's Flow Theory with Juul's ludological framework, we propose the concept of Resilient Flow. We define this as a cognitive state wherein absorption is maintained not despite frustration but through the meaningful framing of it. To validate this model without invasive laboratory constraints, we conducted a qualitative text analysis of 600 helpful user reviews from Elden Ring, Sekiro: Shadows Die Twice, and Dark Souls III via the Steam Community platform. Findings reveal that long-term players linguistically reframe death as pedagogy rather than punishment and utilize vocabulary associated with rhythmic synchronization and meditative focus. We conclude that when difficulty is designed with clarity and fairness, it fosters an Ethics of Attention and transforms digital struggle into a profound experience of mastery and mindfulness.

cs.HC↗

arXiv2026

Universality of extremal eigenvalues of large random matrices

We prove that the spectral radius of a large random matrix $X$ with independent, identically distributed complex entries follows the Gumbel law irrespective of the distribution of the matrix elements. This solves a long-standing conjecture of Bordenave and Chafa{\"ı} and it establishes the first universality result for one of the most prominent extremal spectral statistics in random matrix theory. Furthermore, we also prove that the argument of the largest eigenvalue is uniform on the unit circle and that the extremal eigenvalues of $X$ form a Poisson point process.

math.PR↗

arXiv2026

Optimal decay of eigenvector overlap for non-Hermitian random matrices

We consider the standard overlap $\mathcal{O}_{ij}: =\langle \mathbf{r}_j, \mathbf{r}_i\rangle\langle \mathbf{l}_j, \mathbf{l}_i\rangle$ of any bi-orthogonal family of left and right eigenvectors of a large random matrix $X$ with centred i.i.d. entries and we prove that it decays as an inverse second power of the distance between the corresponding eigenvalues. This extends similar results for the complex Gaussian ensemble from Bourgade and Dubach [arXiv:1801.01219], as well as Benaych-Georges and Zeitouni [arXiv:1806.06806], to any i.i.d. matrix ensemble in both symmetry classes. As a main tool, we prove a two-resolvent local law for the Hermitisation of $X$ uniformly in the spectrum with optimal decay rate and optimal dependence on the density near the spectral edge.

math.PR↗

arXiv2026

Electrical Regulation of Transverse Spin Currents in Unconventional Magnetic Ferroeletrics

We identify hexagonal YMnO$_3$ as a material realization of the elusive $β$-phase of unconventional magnetism, a noncollinear, noncoplanar antiferromagnetic state defined by intrinsic spin-momentum locking and a topological spin texture. First-principle calculations reveal that this unique electronic structure enables a perpendicular electric field to generate a transverse pure spin current, a response that occurs without requiring relativistic spin-orbit coupling. Symmetry analysis demonstrates that this spin current is intimately related to the material's ferroelectric polarization that breaks the inversion symmetry and is rigorously forbidden at domain walls where electrical polarization vanishes. This provides a blueprint for a non-volatile transistor where a gate voltage switches the spin current conductivity by controlling domain wall density, enabling all-electrical control for energy-efficient antiferromagnetic spintronics.

cond-mat.mtrl-sci↗

arXiv2025

Large deviations for the extremal eigenvalues of Ginibre ensembles

We establish large deviation principles for the extremal eigenvalues of the Ginibre ensembles with good rate functions. In contrast to the typical estimates for the extremal eigenvalues, the large deviations for the real Ginibre ensemble come from the eigenvalues lying on the real line. Moreover, we also derive deviation estimates for the second leading term in the asymptotic expansion of the extremal eigenvalues. These polynomially small deviation estimates are universal for any i.i.d. matrices under a mild moment condition.

math.PR↗

arXiv2025

Rethinking Wine Tasting for Chinese Consumers: A Service Design Approach Enhanced by Multimodal Personalization

Wine tasting is a multimodal and culturally embedded activity that presents unique challenges when adapted to non-Western contexts. This paper proposes a service design approach rooted in contextual co-creation to reimagine wine tasting experiences for Chinese consumers. Drawing on 26 in-situ interviews and follow-up validation sessions, we identify three distinct user archetypes: Curious Tasters, Experience Seekers, and Knowledge Builders, each exhibiting different needs in vocabulary, interaction, and emotional pacing. Our findings reveal that traditional wine descriptors lack cultural resonance and that cross-modal metaphors grounded in local gastronomy (e.g., green mango for acidity) significantly improve cognitive and emotional engagement. These insights informed a partially implemented prototype, featuring AI-driven metaphor-to-flavour mappings and real-time affective feedback visualisation. A small-scale usability evaluation confirmed improvements in engagement and comprehension. Our comparative analysis shows alignment with and differentiation from prior multimodal and affect-aware tasting systems. This research contributes to CBMI by demonstrating how culturally adaptive interaction systems can enhance embodied consumption experiences in physical tourism and beyond.

cs.HC↗

arXiv2025

Designing Wine Tasting Experiences for All: The role of Human Diversity and Personal food memory

This study investigates the design of inclusive wine-tasting experiences by examining the roles of human diversity and personal food memory. Through field studies conducted in various wine regions, we explored how Chinese visitors engage with wine-tasting activities during winery tours, highlighting the cross-cultural challenges they face. Our findings underscore the importance of experiencers' abilities, necessities, and aspirations (ANAs), the authenticity of wine tasting within the context of winery tours, and the use of personal food memories as a wine-tasting tool accessible to all. These insights lay the groundwork for developing more inclusive and engaging wine-tasting services, offering new perspectives for cultural exchange and sustainable wine business practices in China.

cs.HC↗

arXiv2025

Unlocking Multi-Modal Potentials for Link Prediction on Dynamic Text-Attributed Graphs

Dynamic Text-Attributed Graphs (DyTAGs) are a novel graph paradigm that captures evolving temporal events (edges) alongside rich textual attributes. Existing studies can be broadly categorized into TGNN-driven and LLM-driven approaches, both of which encode textual attributes and temporal structures for DyTAG representation. We observe that DyTAGs inherently comprise three distinct modalities: temporal, textual, and structural, often exhibiting completely disjoint distributions. However, the first two modalities are largely overlooked by existing studies, leading to suboptimal performance. To address this, we propose MoMent, a multi-modal model that explicitly models, integrates, and aligns each modality to learn node representations for link prediction. Given the disjoint nature of the original modality distributions, we first construct modality-specific features and encode them using individual encoders to capture correlations across temporal patterns, semantic context, and local structures. Each encoder generates modality-specific tokens, which are then fused into comprehensive node representations with a theoretical guarantee. To avoid disjoint subspaces of these heterogeneous modalities, we propose a dual-domain alignment loss that first aligns their distributions globally and then fine-tunes coherence at the instance level. This enhances coherent representations from temporal, textual, and structural views. Extensive experiments across seven datasets show that MoMent achieves up to 17.28% accuracy improvement and up to 31x speed-up against eight baselines.

cs.LG↗

arXiv2025

Quantitative Tracy-Widom laws for sparse random matrices

We consider the fluctuations of the largest eigenvalue of sparse random matrices, the class of random matrices that includes the normalized adjacency matrices of the Erdős-Rényi graph $G(N, p)$. We show that the fluctuations of the largest eigenvalue converge to the Tracy-Widom law at a rate almost $O(N^{-1/3 } + p^{-2} N^{-4/3})$ in the regime $p \gg N^{-2/3 }$. Our proof builds upon the Green function comparison method initiated by Erdős, Yau, and Yin [22]. To show a Green function comparison theorem for fine spectral scales, we implement algorithms for symbolic computations involving averaged products of Green function entries.

math.PR↗

arXiv2025

Probing scattering of Raman phonons on magnetic and electronic excitations in pyrochlores Nd$_2$Zr$_2$O$_7$ and Nd$_2$Ir$_2$O$_7$

Magnetic rare earth atoms on pyrochlore lattice can produce such exotic magnetic states as spin ice and quantum spin ice. These states are a result of the frustration in the pyrochlore lattice, as well as crystal field degrees of freedom of rare earth atoms, and their interactions with the lattice. Raman scattering spectroscopy, which possess high spectral resolution and can easily access broad energy and temperature ranges, is an optimum tool to study these excitations and their interactions. In this work we follow Raman scattering of zone center phonons and crystal field excitations of Nd$^{3+}$ in Nd$_2$Zr$_2$O$_7$ and Nd$_2$Ir$_2$O$_7$ in the temperature range where these materials are paramagnetic. A comparison between an insulating Nd$_2$Zr$_2$O$_7$ and semimetallic Nd$_2$Ir$_2$O$_7$ materials allow us to distinguish between scattering of phonons on other phonons, crystal field excitations, and electrons, highlighting interactions between these degrees of freedom.

cond-mat.str-el↗

arXiv2025

Mentor-Telemachus Bond: Transferring Knowledge in Semantic Communication via Contrastive Learning

Encoder, decoder and knowledge base are three major components for semantic communication. Recent advances have achieved significant progress in the encoder-decoder design. However, there remains a considerable gap in the construction and utilization of knowledge base, which plays important roles in establishing consensus among communication participants through knowledge transferring and sharing. Current knowledge base designs typically involve complex structures, which lead to significant computational overheads and heavy reliance on manually annotated datasets, making it difficult to adapt to existing encoder-decoder models. Hence, without knowledge transferring and sharing within the network results in poor generalization of encoder-decoder. This necessitates model training for specific tasks and datasets, significantly limiting the scalability of semantic communication systems to larger networks. To address these challenges, we propose an innovative Contrastive Representations Learning based Semantic Communication Framework (CRLSC). In CRLSC, the server-side pre-trained large model utilizes large-scale public datasets to construct shared knowledge base. Local-side encoders in terminal devices conduct training guided by shared knowledge base. These trained encoders can then build private knowledge bases from private datasets and fine-tune decoders for specific tasks. This simple and effective approach can facilitate the knowledge transferring across large-scale heterogeneous networks.

cs.NI↗