Searcharxiv⌕ Search

arXiv subjects

Yan Sun

Publications and source records attributed to Yan Sun.

At least 55 records · Page 3Linked to original sources

Multinoulli Extension: A Lossless Continuous Relaxation for Partition-Constrained Subset Selection

Identifying the most representative subset for a close-to-submodular objective while satisfying the predefined partition constraint is a fundamental task with numerous applications in machine learning. However, the existing distorted local-search methods are often hindered by their prohibitive query complexities and the rigid requirement for prior knowledge of difficult-to-obtain structural parameters. To overcome these limitations, we introduce a novel algorithm titled Multinoulli-SCG, which not only is parameter-free, but also can achieve the same approximation guarantees as the distorted local-search methods with significantly fewer function evaluations. More specifically, when the objective function is monotone $α$-weakly DR-submodular or $(γ,β)$-weakly submodular, our Multinoulli-SCG algorithm can attain a value of $(1-e^{-α})\text{OPT}-ε$ or $(\frac{γ^{2}(1-e^{-(β(1-γ)+γ^2)})}{β(1-γ)+γ^2})\text{OPT}-ε$ with only $O(1/ε^{2})$ function evaluations, where OPT denotes the optimal value. The cornerstone of our Multinoulli-SCG algorithm is an innovative continuous-relaxation framework named Multinoulli Extension(ME), which can effectively convert the discrete subset selection problem subject to partition constraints into a solvable continuous maximization focused on learning the optimal multinoulli priors across the concerned partition. In sharp contrast with the well-established multi-linear extension for submodular subset selection, a notable advantage of our proposed ME is its intrinsic capacity to provide a lossless rounding scheme for any set function. Furthermore, based on our proposed ME, we also present two novel online algorithms, namely, Multinoulli-OSCG and Multinoulli-OSGA, for the unexplored online subset selection problems over partition constraints.

cs.LG↗

Time Tracker: Mixture-of-Experts-Enhanced Foundation Time Series Forecasting Model with Decoupled Training Pipelines

In the past few years, time series foundation models have achieved superior predicting accuracy. However, real-world time series often exhibit significant diversity in their temporal patterns across different time spans and domains, making it challenging for a single model architecture to fit all complex scenarios. In addition, time series data may have multiple variables exhibiting complex correlations between each other. Recent mainstream works have focused on modeling times series in a channel-independent manner in both pretraining and finetuning stages, overlooking the valuable inter-series dependencies. To this end, we propose Time Tracker for better predictions on multivariate time series data. Firstly, we leverage sparse mixture of experts (MoE) within Transformers to handle the modeling of diverse time series patterns, thereby alleviating the learning difficulties of a single model while improving its generalization. Besides, we propose Any-variate Attention, enabling a unified model structure to seamlessly handle both univariate and multivariate time series, thereby supporting channel-independent modeling during pretraining and channel-mixed modeling for finetuning.Furthermore, we design a graph learning module that constructs relations among sequences from frequency-domain features, providing more precise guidance to capture inter-series dependencies in channel-mixed modeling. Based on these advancements, Time Tracker achieves state-of-the-art performance in predicting accuracy, model generalization and adaptability.

cs.LG↗

High-pressure phase stability and superconductivity in La-Zr-H hydrides

Hydrogen-rich ternary hydrides are promising candidates for high-Tc superconductivity at megabar pressures, yet their chemical space is vast and largely unexplored. Combining evolutionary structure searches with first-principles calculations, we comprehensively investigate the La-Zr-H ternary system in the 150-300 GPa pressure range. Zero-point energy-corrected convex hull analysis identifies multiple stable superconducting phases, including R3m-Zr2H17 at 300 GPa and P6/mmm-LaZr2H24 at 200 GPa, both of which are thermodynamically and dynamically stable and exhibit strong electron-phonon coupling. Solution of the Eliashberg equations predicts high superconducting transition temperatures of Tc = 209 K for R3m-Zr2H17 at 300 GPa and Tc = 202 K for P6/mmm-LaZr2H24 at 200 GPa. In addition to these stable phases, we identify a high-symmetry metastable compound, P6m2-LaZrH18, which lies just 0.027 eV/atom above the convex hull yet remains dynamically stable and exhibits a high predicted Tc of 206 K at 300 GPa. We find that, across all phases, the elevated Tc correlates with the high-symmetry structure with dense hydrogen cages, favorable electron counts per hydrogen, and a large hydrogen-derived density of states at the Fermi level. Finally, a random- forest machine learning model, trained on diverse hydrides superconductivity data, reproduces these structure-property trends across predicted structures, enabling to identify potential hydrides with high predicted Tc for targeted follow-up calculations and future high-pressure experiments.

cond-mat.mtrl-sci↗

VLBI astrometry of radio stars to link radio and optical celestial reference frames - II. 11 radio stars

The alignment between the radio-based International Celestial Reference Frame (ICRF) and the optical Gaia Celestial Reference Frame (Gaia-CRF) is critical for multi-waveband astronomy, yet systematic offsets at the optical bright end (G<13) limit their consistency. While radio stars offer a potential link between these frames, their utility has been restricted by the scarcity of precise Very Long Baseline Interferometry (VLBI) astrometry. In this study, we present new VLBI astrometry of 11 radio stars using the Very Long Baseline Array (VLBA), expanding the existing sample with positions, parallaxes, and proper motions measured. All 11 radio stars were detected, for 10 of which parallaxes and proper motions can be estimated, achieving median uncertainties better than 0.1 mas and 0.1 mas/yr, respectively. These new samples greatly contribute to the link between ICRF and Gaia-CRF at the optical bright end.

astro-ph.SR↗

Entropy-Guided Dynamic Tokens for Graph-LLM Alignment in Molecular Understanding

Molecular understanding is central to advancing areas such as scientific discovery, yet Large Language Models (LLMs) struggle to understand molecular graphs effectively. Existing graph-LLM bridges often adapt the Q-Former-style connector with fixed-length static tokens, which is originally designed for vision tasks. These designs overlook stereochemistry and substructural context and typically require costly LLM-backbone fine-tuning, limiting efficiency and generalization. We introduce EDT-Former, an Entropy-guided Dynamic Token Transformer that generates tokens aligned with informative molecular patches, thereby preserving both local and global structural features for molecular graph understanding. Beyond prior approaches, EDT-Former enables alignment between frozen graph encoders and LLMs without tuning the LLM backbone (excluding the embedding layer), resulting in computationally efficient finetuning, and achieves stateof-the-art results on MoleculeQA, Molecule-oriented Mol-Instructions, and property prediction benchmarks (TDC, MoleculeNet), underscoring its effectiveness for scalable and generalizable multimodal molecular understanding

cs.LG↗

Stability and Generalization of Push-Sum Based Decentralized Optimization over Directed Graphs

Push-Sum-based decentralized learning enables optimization over directed communication networks, where information exchange may be asymmetric. While convergence properties of such methods are well understood, their finite-iteration stability and generalization behavior remain unclear due to structural bias induced by column-stochastic mixing and asymmetric error propagation. In this work, we develop a unified uniform-stability framework for the Stochastic Gradient Push (SGP) algorithm that captures the effect of directed topology. A key technical ingredient is an imbalance-aware consistency bound for Push-Sum, which controls consensus deviation through two quantities: the stationary distribution imbalance parameter $δ$ and the spectral gap $(1-λ)$ governing mixing speed. This decomposition enables us to disentangle statistical effects from topology-induced bias. We establish finite-iteration stability and optimization guarantees for both convex objectives and non-convex objectives satisfying the Polyak--Łojasiewicz condition. For convex problems, SGP attains excess generalization error of order $\tilde{\mathcal{O}}\!\left(\frac{1}{\sqrt{mn}}+\fracγ{δ(1-λ)}+γ\right)$ under step-size schedules, and we characterize the corresponding optimal early stopping time that minimizes this bound. For PŁ objectives, we obtain convex-like optimization and generalization rates with dominant dependence proportional to $κ\!\left(1+\frac{1}{δ(1-λ)}\right)$, revealing a multiplicative coupling between problem conditioning and directed communication topology. Our analysis clarifies when Push-Sum correction is necessary compared with standard decentralized SGD and quantifies how imbalance and mixing jointly shape the best attainable learning performance.

cs.LG↗

Foundations of Top-$k$ Decoding For Language Models

Top-$k$ decoding is a widely used method for sampling from LLMs: at each token, only the largest $k$ next-token-probabilities are kept, and the next token is sampled after re-normalizing them to sum to unity. Top-$k$ and other sampling methods are motivated by the intuition that true next-token distributions are sparse, and the noisy LLM probabilities need to be truncated. However, to our knowledge, a precise theoretical motivation for the use of top-$k$ decoding is missing. In this work, we develop a theoretical framework that both explains and generalizes top-$k$ decoding. We view decoding at a fixed token as the recovery of a sparse probability distribution. We consider \emph{Bregman decoders} obtained by minimizing a separable Bregman divergence (for both the \emph{primal} and \emph{dual} cases) with a sparsity-inducing $\ell_0$ regularization. Despite the combinatorial nature of the objective, we show how to optimize it efficiently for a large class of divergences. We show that the optimal decoding strategies are greedy, and further that the loss function is discretely convex in $k$, so that binary search provably and efficiently finds the optimal $k$. We show that top-$k$ decoding arises as a special case for the KL divergence, and identify new decoding strategies that have distinct behaviors (e.g., non-linearly up-weighting larger probabilities after re-normalization).

cs.AI↗

Distilling and Adapting: A Topology-Aware Framework for Zero-Shot Interaction Prediction in Multiplex Biological Networks

Multiplex Biological Networks (MBNs), which represent multiple interaction types between entities, are crucial for understanding complex biological systems. Yet, existing methods often inadequately model multiplexity, struggle to integrate structural and sequence information, and face difficulties in zero-shot prediction for unseen entities with no prior neighbourhood information. To address these limitations, we propose a novel framework for zero-shot interaction prediction in MBNs by leveraging context-aware representation learning and knowledge distillation. Our approach leverages domain-specific foundation models to generate enriched embeddings, introduces a topology-aware graph tokenizer to capture multiplexity and higher-order connectivity, and employs contrastive learning to align embeddings across modalities. A teacher-student distillation strategy further enables robust zero-shot generalization. Experimental results demonstrate that our framework outperforms state-of-the-art methods in interaction prediction for MBNs, providing a powerful tool for exploring various biological interactions and advancing personalized therapeutics.

cs.LG↗

Efficient photocatalytic CO2 Reduction to C2+ Products with Pt1-xPdxSn4 Dirac Nodal Arc Semimetal

The photochemical CO2 reduction reaction represents a zero-carbon pathway for converting CO2 into value-added chemicals, yet its industrial implementation has been constrained by low selectivity and product diversity. Dirac nodal arc semimetals characterized by ultrahigh carrier mobility with over 25000 cm2 V-1 s-1 offer a promising platform to search for efficient catalysts for CO2 conversion. Herein, we demonstrate that strategic Pt incorporation into PdSn4 optimizes the electronic structure and carrier dynamics of this Dirac semimetal. Experimental and theoretical analyses reveal that the resulting Pd-Sn-Pt local electronic structure redistributes charge density around Pd and Pt atoms, which facilitates C-C coupling via *OC-COH and *OC-CHOH intermediates and enhances carrier mobility by 40% versus the pristine PdSn4 single crystal. The optimized Pd0.4Pt0.6Sn4 single crystal achieves C2H4 with formation rate of 0.000328 mol g-1 h-1, product selectivity of 73.1% and electron-based selectivity of 89%. This work establishes electronic-structure-tunable Dirac semimetals as a new paradigm for multi-carbon photochemical CO2 reduction, providing a design strategy for next-generation photocatalysts.

cond-mat.mtrl-sci↗

Efficient Reinforcement Learning for Large Language Models with Intrinsic Exploration

Reinforcement learning with verifiable rewards (RLVR) has improved the reasoning ability of large language models, yet training remains costly because many rollouts contribute little to optimization, considering the amount of computation required. This study investigates how simply leveraging intrinsic data properties, almost free benefit during training, can improve data efficiency for RLVR. We propose PREPO with two complementary components. First, we adopt prompt perplexity as an indicator of model adaptability in learning, enabling the model to progress from well-understood contexts to more challenging ones. Second, we amplify the discrepancy among the rollouts by differentiating their relative entropy, and prioritize sequences that exhibit a higher degree of exploration. Together, these mechanisms reduce rollout demand while preserving competitive performance. On the Qwen and Llama models, PREPO achieves effective results on mathematical reasoning benchmarks with up to 3 times fewer rollouts than the baselines. Beyond empirical gains, we provide theoretical and in-depth analyses explaining the underlying rationale of our method to improve the data efficiency of RLVR.

cs.LG↗

The distinction of time-reversal-like degeneracy by electronic transport in a new compound

We report the discovery of a new compound, Ce$_3$MgBi$_5$, and reveal the hidden time-reversal-like degenerate states within it. Ce$_3$MgBi$_5$ is an antiferromagnet with the distorted kagome lattice of Ce atoms, in which several fractional magnetization plateaus emerge with the increase of magnetic field. At the 1/2 magnetization plateau, obvious hysteresis has been observed in the magnetoresistance and Hall resistivity during the rise and fall of the magnetic field. However, hysteresis vanishes in the corresponding measurements of magnetization, indicating the existence of degenerate states with the same net magnetization but different electronic transport properties. The degenerate states can be connected by the time-reversal-like operation. In addition, by comparing with HoAgGe, it is suggested that the special crystal structure in Ce$_3$MgBi$_5$ may have a shielding effect on the time-reversal-like operation, thereby affecting the distinction of degenerate states. Our work establishes Ce$_3$MgBi$_5$ as an example of utilizing electronic transport properties to identify and distinguish hidden symmetries in frustrated magnetic systems.

cond-mat.str-el↗

CATS: Enhancing Multivariate Time Series Forecasting by Constructing Auxiliary Time Series as Exogenous Variables

For Multivariate Time Series Forecasting (MTSF), recent deep learning applications show that univariate models frequently outperform multivariate ones. To address the difficiency in multivariate models, we introduce a method to Construct Auxiliary Time Series (CATS) that functions like a 2D temporal-contextual attention mechanism, which generates Auxiliary Time Series (ATS) from Original Time Series (OTS) to effectively represent and incorporate inter-series relationships for forecasting. Key principles of ATS - continuity, sparsity, and variability - are identified and implemented through different modules. Even with a basic 2-layer MLP as core predictor, CATS achieves state-of-the-art, significantly reducing complexity and parameters compared to previous multivariate models, marking it an efficient and transferable MTSF solution.

stat.ML↗

In-context Time Series Predictor

Recent Transformer-based large language models (LLMs) demonstrate in-context learning ability to perform various functions based solely on the provided context, without updating model parameters. To fully utilize the in-context capabilities in time series forecasting (TSF) problems, unlike previous Transformer-based or LLM-based time series forecasting methods, we reformulate "time series forecasting tasks" as input tokens by constructing a series of (lookback, future) pairs within the tokens. This method aligns more closely with the inherent in-context mechanisms, and is more parameter-efficient without the need of using pre-trained LLM parameters. Furthermore, it addresses issues such as overfitting in existing Transformer-based TSF models, consistently achieving better performance across full-data, few-shot, and zero-shot settings compared to previous architectures.

cs.LG↗

WAVE: Weighted Autoregressive Varying Gate for Time Series Forecasting

We propose a Weighted Autoregressive Varying gatE (WAVE) attention mechanism equipped with both Autoregressive (AR) and Moving-average (MA) components. It can adapt to various attention mechanisms, enhancing and decoupling their ability to capture long-range and local temporal patterns in time series data. In this paper, we first demonstrate that, for the time series forecasting (TSF) task, the previously overlooked decoder-only autoregressive Transformer model can achieve results comparable to the best baselines when appropriate tokenization and training methods are applied. Moreover, inspired by the ARMA model from statistics and recent advances in linear attention, we introduce the full ARMA structure into existing autoregressive attention mechanisms. By using an indirect MA weight generation method, we incorporate the MA term while maintaining the time complexity and parameter size of the underlying efficient attention models. We further explore how indirect parameter generation can produce implicit MA weights that align with the modeling requirements for local temporal impacts. Experimental results show that WAVE attention that incorporates the ARMA structure consistently improves the performance of various AR attentions on TSF tasks, achieving state-of-the-art results.

cs.LG↗

ZeroS: Zero-Sum Linear Attention for Efficient Transformers

Linear attention methods offer Transformers $O(N)$ complexity but typically underperform standard softmax attention. We identify two fundamental limitations affecting these approaches: the restriction to convex combinations that only permits additive information blending, and uniform accumulated weight bias that dilutes attention in long contexts. We propose Zero-Sum Linear Attention (ZeroS), which addresses these limitations by removing the constant zero-order term $1/t$ and reweighting the remaining zero-sum softmax residuals. This modification creates mathematically stable weights, enabling both positive and negative values and allowing a single attention layer to perform contrastive operations. While maintaining $O(N)$ complexity, ZeroS theoretically expands the set of representable functions compared to convex combinations. Empirically, it matches or exceeds standard softmax attention across various sequence modeling benchmarks.

cs.LG↗

Singleton-Optimized Conformal Prediction

Conformal prediction can be used to construct prediction sets that cover the true outcome with a desired probability, but can sometimes lead to large prediction sets that are costly in practice. The most useful outcome is a singleton prediction-an unambiguous decision-yet existing efficiency-oriented methods primarily optimize average set size. Motivated by this, we propose a new nonconformity score that aims to minimize the probability of producing non-singleton sets. Starting from a non-convex constrained optimization problem as a motivation, we provide a geometric reformulation and associated algorithm for computing the nonconformity score and associated split conformal prediction sets in O(K) time for K-class problems. Using this score in split conformal prediction leads to our proposed Singleton-Optimized Conformal Prediction (SOCOP) method. We evaluate our method in experiments on image classification and LLM multiple-choice question-answering, comparing with standard nonconformity scores such as the (negative) label probability estimates and their cumulative distribution function; both of which are motivated by optimizing length. The results show that SOCOP increases singleton frequency (sometimes by over 20%) compared to the above scores, with minimal impact on average set size.

stat.ML↗

How is Cold Gas Loaded into Galactic Nuclear Outflows?

The origin of the multiphase gas within the Fermi/eROSITA bubbles is crucial for understanding Galactic Center (GC) feedback. We use HI4PI data to investigate the kinematics and physical properties of high-velocity clouds (HVCs) toward the GC. Our results reveal that the HVCs exhibit a distinct asymmetric distribution, closely associated with the bar-driven tilted dust lanes and the distorted overshooting streams. We propose that powerful nuclear outflows interact with these gas-rich, off-plane structures, striping and entraining cold gas from the outer Galactic regions (R_GC~0.5--1.7 kpc) rather than solely from the region of the central molecular zone (CMZ; R_GC<0.3 kpc). In this scenario, as the Galactic bar drives gas inflows along the dust lanes, nuclear outflows simultaneously break through the CMZ, sweeping up and ablating cold gas from the boundary layer of these pre-existing structures. This process naturally accounts for the observed high turbulence, complex spectral signatures, and anomalous spatial-kinematic gas patterns, as well as multiwavelength asymmetries of the bubbles. The HVCs are accelerated to about 230--340 km/s over a dynamical time of ~3--6 Myr. When the multiphase, inhomogeneous composition of the gas is included, the estimated gas outflow rate reaches ~1 Msun/yr. This value is comparable to the bar-driven inflow rate, indicating a tightly coupled gas cycle in the inner Galaxy. Our research highlights the critical role of bar-driven gas dynamics and nuclear feedback in the secular evolution of the Milky Way, offering a valuable paradigm for investigating gas cycles in external galaxies.

astro-ph.GA↗

Complex spin dynamics induced metamagnetic phase transitions in Dirac semimetal EuAuBi

We report a comprehensive investigation of the physical properties of the Dirac semimetal compound EuAuBi single crystals, using neutron diffraction, magnetization, electrical transport, and specific heat measurements. EuAuBi crystallizes in a hexagonal structure with space group P63mc (No. 186). First-principles calculations using density functional theory characterize it as a Dirac semimetal, with a notable band-crossing in proximity to the Fermi level (EF ) along the Γ-A direction. The crystal exhibits three distinct magnetic phases at 4 K (TN1), 3.5 K (TN2), and 2.8 K (TN3)as observed from magnetic and specific heat measurements. However, zero-field neutron diffraction resolves only two magnetic phases: a commensurate antiferromagnetic phase and a canted antiferromagnetic phase. Field-dependent ac and dc magnetization measurements uncover field-induced non-trivial spin textures in the magnetic field range 1.5 to 3 T, manifested as a tilted plateau in the magnetization curves. The interplay between conduction carriers and these spin textures is further evidenced by unique features in the magnetic field-dependent longitudinal resistivity in the system. Finally, we present a comprehensive magnetic phase diagram of EuAuBi, highlighting diverse spin alignments present in the material. EuAuBi thus emerges as a rare material system in which both momentum-space and real-space Berry curvature effects may coexist, providing a unique opportunity to investigate their interplay.

cond-mat.mtrl-sci↗