Search

Qualitative structure-activity relationship (QSAR) is important for drug discovery and offers valuable insights into the biological interactions of potential drug candidates. It has been demonstrated that QSAR can be accurately predicted by machine learning. However, data with poor quality and limited availability are always the most common and critical issues for medical-related applications for machine learning. In this manuscript, we aim to discuss the performance of classical and quantum classifiers in QSAR prediction and attempt to demonstrate the quantum advantages in the generalization power of the quantum classifier under conditions of limited data availability and a reduced number of features. By applying different data embedding methods followed by feature selection through principal component analysis (PCA), we find that the quantum classifier outperforms the classical one when a small number of features are selected and the number of training samples is limited. The generality of quantum advantages in other open datasets is also explored.

This roadmap consolidates recent advances while exploring emerging applications, reflecting the remarkable diversity of hardware platforms, neuromorphic concepts, and implementation philosophies reported in the field. It emphasizes the critical role of cross-disciplinary collaboration in this rapidly evolving field.

In the relentless pursuit of quantum computational advantage, we present a significant advancement with the development of Zuchongzhi 3.0. This superconducting quantum computer prototype, comprising 105 qubits, achieves high operational fidelities, with single-qubit gates, two-qubit gates, and readout fidelity at 99.90%, 99.62% and 99.18%, respectively. Our experiments with an 83-qubit, 32-cycle random circuit sampling on Zuchongzhi 3.0 highlight its superior performance, achieving one million samples in just a few hundred seconds. This task is estimated to be infeasible on the most powerful classical supercomputers, Frontier, which would require approximately $6.4\times 10^9$ years to replicate the task. This leap in processing power places the classical simulation cost six orders of magnitude beyond Google's SYC-67 and SYC-70 experiments [Nature 634, 328(2024)], firmly establishing a new benchmark in quantum computational advantage. Our work not only advances the frontiers of quantum computing but also lays the groundwork for a new era where quantum processors play an essential role in tackling sophisticated real-world challenges.

Functionalities in crystalline materials are determined by 3-dimensional collective interactions of atoms. The confinement of dimensionality in condensed matter provides an exotic research direction to understand the interaction of atoms, thus can be used to tailor or create new functionalities in material systems. In this study, a 2-dimensional transition metal oxide monolayer is constructed inside complex oxide heterostructures based on the theoretical predictions. The electrostatic boundary conditions of oxide monolayer in the heterostructure is carefully designed to tune the chemical, electronic, and magnetic states of oxide monolayer. The challenge of characterizing such an oxide monolayer is overcome by a combination of transmission electron microscopy, x-ray absorption spectroscopy, cross-sectional scanning tunneling microscopy, and electrical transport measurements. An intriguing metal-insulator transition associated with a magnetic transition is discovered in the MnO2 monolayer. This study paves a new route to understand the confinement of dimensionality and explore new intriguing phenomena in condensed matters., Comment: 23 pages, 6 figures

Equivariant Graph Neural Networks (EGNNs) have demonstrated significant success in modeling microscale systems, including those in chemistry, biology and materials science. However, EGNNs face substantial computational challenges due to the high cost of constructing edge features via spherical tensor products, making them impractical for large-scale systems. To address this limitation, we introduce E2Former, an equivariant and efficient transformer architecture that incorporates the Wigner $6j$ convolution (Wigner $6j$ Conv). By shifting the computational burden from edges to nodes, the Wigner $6j$ Conv reduces the complexity from $O(|\mathcal{E}|)$ to $ O(| \mathcal{V}|)$ while preserving both the model's expressive power and rotational equivariance. We show that this approach achieves a 7x-30x speedup compared to conventional $\mathrm{SO}(3)$ convolutions. Furthermore, our empirical results demonstrate that the derived E2Former mitigates the computational challenges of existing approaches without compromising the ability to capture detailed geometric information. This development could suggest a promising direction for scalable and efficient molecular modeling.

We present a novel approach centered on the decoding stage of Automatic Speech Recognition (ASR) that enhances multilingual performance, especially for low-resource languages. It utilizes a cross-lingual embedding clustering method to construct a hierarchical Softmax (H-Softmax) decoder, which enables similar tokens across different languages to share similar decoder representations. It addresses the limitations of the previous Huffman-based H-Softmax method, which relied on shallow features in token similarity assessments. Through experiments on a downsampled dataset of 15 languages, we demonstrate the effectiveness of our approach in improving low-resource multilingual ASR accuracy.

Supervised fine-tuning (SFT) and reinforcement learning (RL) are widely used post-training techniques for foundation models. However, their roles in enhancing model generalization capabilities remain unclear. This paper studies the difference between SFT and RL on generalization and memorization, focusing on text-based rule variants and visual variants. We introduce GeneralPoints, an arithmetic reasoning card game, and adopt V-IRL, a real-world navigation environment, to assess how models trained with SFT and RL generalize to unseen variants in both textual and visual domains. We show that RL, especially when trained with an outcome-based reward, generalizes across both rule-based textual and visual variants. SFT, in contrast, tends to memorize training data and struggles to generalize out-of-distribution scenarios. Further analysis reveals that RL improves the model's underlying visual recognition capabilities, contributing to its enhanced generalization in the visual domain. Despite RL's superior generalization, we show that SFT remains essential for effective RL training; SFT stabilizes the model's output format, enabling subsequent RL to achieve its performance gains. These findings demonstrates the capability of RL for acquiring generalizable knowledge in complex, multi-modal tasks., Comment: Website at https://tianzhechu.com/SFTvsRL

Three-dimensional (3D) multi-slab acquisition is a technique frequently employed in high-resolution diffusion-weighted MRI in order to achieve the best signal-to-noise ratio (SNR) efficiency. However, this technique is limited by slab boundary artifacts that cause intensity fluctuations and aliasing between slabs which reduces the accuracy of anatomical imaging. Addressing this issue is crucial for advancing diffusion MRI quality and making high-resolution imaging more feasible for clinical and research applications. In this work, we propose a regularized slab profile encoding (PEN) method within a Plug-and-Play ADMM framework, incorporating multi-scale energy (MuSE) regularization to effectively improve the slab combined reconstruction. Experimental results demonstrate that the proposed method significantly improves image quality compared to non-regularized and TV-regularized PEN approaches. The regularized PEN framework provides a more robust and efficient solution for high-resolution 3D diffusion MRI, potentially enabling clearer, more reliable anatomical imaging across various applications., Comment: 4 pages, 4 figures, ISBI2025 Conference paper

Sterile neutrinos with masses on the $\mathrm{eV}$ scale are promising candidates to account for the origin of neutrino mass and the reactor neutrino anomalies. The mixing between sterile and active neutrinos in the early universe could result in a large abundance of relic sterile neutrinos, which depends on not only their physical mass $m_{\rm phy}$ but also their degree of thermalization, characterized by the extra effective number of relativistic degrees of freedom $\Delta N_{\rm eff}$. Using neutrino-involved N-body simulations, we investigate the effects of sterile neutrinos on the matter power spectrum, halo pairwise velocity, and halo mass and velocity functions. We find that the presence of sterile neutrinos suppress the matter power spectrum and halo mass and velocity functions, but enhance the halo pairwise velocity. We also provide fitting formulae to quantify these effects., Comment: 22 pages, 9 main + 2 appendix figures. Second update only corrects the abstract on arXiv page

Quantum computational advantage is a critical milestone for near-term quantum technologies and an essential step towards building practical quantum computers. Recent successful demonstrations of quantum computational advantage owe much to specifically designed random quantum circuit (RQC) protocols that enable hardware-friendly implementation and, more importantly, pose great challenges for classical simulation. Here, we report the automated protocol design approach used for finding the optimal RQC in the \textit{Zuchongzhi} quantum computational advantage experiment [Phys. Rev. Lett. 127 (18), 180501 (2021)]. Without a carefully designed protocol, the classical simulation cost of the \textit{Zuchongzhi}'s 56-qubit 20-cycle RQC experiment would not be considerably higher than Google's 53-qubit 20-cycle experiment, even though more qubits are involved. For Google's latest RQC experiment using $70$ qubits and $24$ cycles [arXiv:2304.11119 (2023)], we estimate that its classical simulation cost can be increased by at least one order of magnitude using our approach. The proposed method can be applied to generic planar quantum processor architectures and addresses realistic imperfections such as processor defects, underpinning quantum computational advantage experiments in future generations of quantum processors., Comment: Intell. Comput. 3:0079 (2024)

The analysis of extended video content poses unique challenges in artificial intelligence, particularly when dealing with the complexity of tracking and understanding visual elements across time. Current methodologies that process video frames sequentially struggle to maintain coherent tracking of objects, especially when these objects temporarily vanish and later reappear in the footage. A critical limitation of these approaches is their inability to effectively identify crucial moments in the video, largely due to their limited grasp of temporal relationships. To overcome these obstacles, we present GraphVideoAgent, a cutting-edge system that leverages the power of graph-based object tracking in conjunction with large language model capabilities. At its core, our framework employs a dynamic graph structure that maps and monitors the evolving relationships between visual entities throughout the video sequence. This innovative approach enables more nuanced understanding of how objects interact and transform over time, facilitating improved frame selection through comprehensive contextual awareness. Our approach demonstrates remarkable effectiveness when tested against industry benchmarks. In evaluations on the EgoSchema dataset, GraphVideoAgent achieved a 2.2 improvement over existing methods while requiring analysis of only 8.2 frames on average. Similarly, testing on the NExT-QA benchmark yielded a 2.0 performance increase with an average frame requirement of 8.1. These results underscore the efficiency of our graph-guided methodology in enhancing both accuracy and computational performance in long-form video understanding tasks.

Recent breakthroughs in large language models (LLMs) exemplified by the impressive mathematical and scientific reasoning capabilities of the o1 model have spotlighted the critical importance of high-quality training data in advancing LLM performance across STEM disciplines. While the mathematics community has benefited from a growing body of curated datasets, the scientific domain at the higher education level has long suffered from a scarcity of comparable resources. To address this gap, we present SCP-116K, a new large-scale dataset of 116,756 high-quality problem-solution pairs, automatically extracted from heterogeneous sources using a streamlined and highly generalizable pipeline. Our approach involves stringent filtering to ensure the scientific rigor and educational level of the extracted materials, while maintaining adaptability for future expansions or domain transfers. By openly releasing both the dataset and the extraction pipeline, we seek to foster research on scientific reasoning, enable comprehensive performance evaluations of new LLMs, and lower the barrier to replicating the successes of advanced models like o1 in the broader science community. We believe SCP-116K will serve as a critical resource, catalyzing progress in high-level scientific reasoning tasks and promoting further innovations in LLM development. The dataset and code are publicly available at https://github.com/AQA6666/SCP-116K-open., Comment: 9 pages, 1 figures

The distributionally robust optimization (DRO)-based graph neural network methods improve recommendation systems' out-of-distribution (OOD) generalization by optimizing the model's worst-case performance. However, these studies fail to consider the impact of noisy samples in the training data, which results in diminished generalization capabilities and lower accuracy. Through experimental and theoretical analysis, this paper reveals that current DRO-based graph recommendation methods assign greater weight to noise distribution, leading to model parameter learning being dominated by it. When the model overly focuses on fitting noise samples in the training data, it may learn irrelevant or meaningless features that cannot be generalized to OOD data. To address this challenge, we design a Distributionally Robust Graph model for OOD recommendation (DRGO). Specifically, our method first employs a simple and effective diffusion paradigm to alleviate the noisy effect in the latent space. Additionally, an entropy regularization term is introduced in the DRO objective function to avoid extreme sample weights in the worst-case distribution. Finally, we provide a theoretical proof of the generalization error bound of DRGO as well as a theoretical analysis of how our approach mitigates noisy sample effects, which helps to better understand the proposed framework from a theoretical perspective. We conduct extensive experiments on four datasets to evaluate the effectiveness of our framework against three typical distribution shifts, and the results demonstrate its superiority in both independently and identically distributed distributions (IID) and OOD., Comment: 14 pages, Accepted by WWW'25

Using $\psi(3686)\rightarrow \pi^{0} h_{c}$ decays from a data sample of $(27.12\pm0.14)\times10^{8}$ $\psi(3686)$ events collected by the BESIII detector at the BEPCII collider, $h_c$ radiative decays to $\gamma\pi^{+}\pi^{-},~\gamma\pi^{+}\pi^{-}\eta,~\gamma2(\pi^{+}\pi^{-})$, and $\gamma p\bar{p}$ are observed for the first time, each with a significance greater than $5\sigma$. The corresponding branching fractions are measured. Furthermore, intermediate states below 2.8 GeV/$c^{2}$ are investigated, leading to the first observation of the decay process of $h_c\rightarrow\gamma f_{2}(1270)\rightarrow\gamma\pi^{+}\pi^{-}$ with a significance of $5.5\,\sigma$. This observation represents the first instance of $h_c$ radiative decay to a tensor state.

As the conventional scaling of logic devices comes to an end, functional wafer backside and 3D transistor stacking are consensus for next-generation logic technology, offering considerable design space extension for powers, signals or even devices on the wafer backside. The Flip FET (FFET), a novel transistor architecture combining 3D transistor stacking and fully functional wafer backside, was recently proposed. With symmetric dual-sided standard cell design, the FFET can deliver around 12.5% cell area scaling and faster but more energy-efficient libraries beyond other stacked transistor technologies such as CFET. Besides, thanks to the novel cell design with dual-sided pins, the FFET supports dual-sided signal routing, delivering better routability and larger backside design space. In this work, we demonstrated a comprehensive FFET evaluation framework considering physical implementation and block-level power-performance-area (PPA) assessment for the first time, in which key functions are dual-sided routing and dual-sided RC extraction. A 32-bit RISC-V core was used for the evaluation here. Compared to the CFET with single-sided signals, the FFET with single-sided signals achieved 23.3% post-P&R core area reduction, 25.0% higher frequency and 11.9% lower power at the same utilization, and 16.0 % higher frequency at the same core area. Meanwhile, the FFET supports dual-sided signals, which can further benefit more from flexible allocation of cell input pins on both sides. By optimizing the input pin density and BEOL routing layer number on each side, 10.6% frequency gain was realized without power degradation compared to the one with single-sided signal routing. Moreover, the routability and power efficiency of FFET barely degrades even with the routing layer number reduced from 12 to 5 on each side, validating the great space for cost-friendly design enabled by FFET., Comment: Accepted by DATE 2025

The narrow bandgap of semiconductors allows for thick, uniform Josephson junction barriers, potentially enabling reproducible, stable, and compact superconducting qubits. We study vertically stacked van der Waals Josephson junctions with semiconducting weak links, whose crystalline structures and clean interfaces offer a promising platform for quantum devices. We observe robust Josephson coupling across 2--12 nm (3--18 atomic layers) of semiconducting WSe$_2$ and, notably, a crossover from proximity- to tunneling-type behavior with increasing weak link thickness. Building on these results, we fabricate a prototype all-crystalline merged-element transmon qubit with transmon frequency and anharmonicity closely matching design parameters. We demonstrate dispersive coupling between this transmon and a microwave resonator, highlighting the potential of crystalline superconductor-semiconductor structures for compact, tailored superconducting quantum devices.

This paper introduces the open-source dataset WanJuanSiLu, designed to provide high-quality training corpora for low-resource languages, thereby advancing the research and development of multilingual models. To achieve this, we have developed a systematic data processing framework tailored for low-resource languages. This framework encompasses key stages such as data extraction, corpus cleaning, content deduplication, security filtering, quality evaluation, and theme classification. Through the implementation of this framework, we have significantly improved both the quality and security of the dataset, while maintaining its linguistic diversity. As of now, data for all five languages have been fully open-sourced. The dataset can be accessed at https://opendatalab.com/applyMultilingualCorpus, and GitHub repository is available at https://github.com/opendatalab/WanJuan3.0

Large Language Models (LLMs) are widely applied to downstream domains. However, current LLMs for high-stakes domain tasks, such as financial investment and legal QA, typically generate brief answers without reasoning processes and explanations. This limits users' confidence in making decisions based on their responses. While original CoT shows promise, it lacks self-correction mechanisms during reasoning. This work introduces Domain$o1$s, which enhances LLMs' reasoning capabilities on domain tasks through supervised fine-tuning and tree search. We construct CoT-stock-2k and CoT-legal-2k datasets for fine-tuning models that activate domain-specific reasoning steps based on their judgment. Additionally, we propose Selective Tree Exploration to spontaneously explore solution spaces and sample optimal reasoning paths to improve performance. We also introduce PROOF-Score, a new metric for evaluating domain models' explainability, complementing traditional accuracy metrics with richer assessment dimensions. Extensive experiments on stock investment recommendation and legal reasoning QA tasks demonstrate Domaino1s's leading performance and explainability. Our code is available at https://anonymous.4open.science/r/Domaino1s-006F/.

The success of Large Language Models (LLMs) in various domains has led researchers to apply them to graph-related problems by converting graph data into natural language text. However, unlike graph data, natural language inherently has sequential order. We observe that when the order of nodes or edges in the natural language description of a graph is shuffled, despite describing the same graph, model performance fluctuates between high performance and random guessing. Additionally, due to the limited input context length of LLMs, current methods typically randomly sample neighbors of target nodes as representatives of their neighborhood, which may not always be effective for accurate reasoning. To address these gaps, we introduce GraphBC. This novel model framework features an Order Selector Module to ensure proper serialization order of the graph and a Subgraph Sampling Module to sample subgraphs with better structure for better reasoning. Furthermore, we propose Graph CoT obtained through distillation, and enhance LLM's reasoning and zero-shot learning capabilities for graph tasks through instruction tuning. Experiments on multiple datasets for node classification and graph question-answering demonstrate that GraphBC improves LLMs' performance and generalization ability on graph tasks.

We introduce DRESS, a novel approach for generating stylized large language model (LLM) responses through representation editing. Existing methods like prompting and fine-tuning are either insufficient for complex style adaptation or computationally expensive, particularly in tasks like NPC creation or character role-playing. Our approach leverages the over-parameterized nature of LLMs to disentangle a style-relevant subspace within the model's representation space to conduct representation editing, ensuring a minimal impact on the original semantics. By applying adaptive editing strengths, we dynamically adjust the steering vectors in the style subspace to maintain both stylistic fidelity and semantic integrity. We develop two stylized QA benchmark datasets to validate the effectiveness of DRESS, and the results demonstrate significant improvements compared to baseline methods such as prompting and ITI. In short, DRESS is a lightweight, train-free solution for enhancing LLMs with flexible and effective style control, making it particularly useful for developing stylized conversational agents. Codes and benchmark datasets are available at https://github.com/ArthurLeoM/DRESS-LLM., Comment: ICLR 2025 Accepted

Benchmarks are important tools for tracking the rapid advancements in large language model (LLM) capabilities. However, benchmarks are not keeping pace in difficulty: LLMs now achieve over 90\% accuracy on popular benchmarks like MMLU, limiting informed measurement of state-of-the-art LLM capabilities. In response, we introduce Humanity's Last Exam (HLE), a multi-modal benchmark at the frontier of human knowledge, designed to be the final closed-ended academic benchmark of its kind with broad subject coverage. HLE consists of 3,000 questions across dozens of subjects, including mathematics, humanities, and the natural sciences. HLE is developed globally by subject-matter experts and consists of multiple-choice and short-answer questions suitable for automated grading. Each question has a known solution that is unambiguous and easily verifiable, but cannot be quickly answered via internet retrieval. State-of-the-art LLMs demonstrate low accuracy and calibration on HLE, highlighting a significant gap between current LLM capabilities and the expert human frontier on closed-ended academic questions. To inform research and policymaking upon a clear understanding of model capabilities, we publicly release HLE at https://lastexam.ai., Comment: 25 pages, 6 figures

Quantum Approximate Optimization Algorithm (QAOA) is a promising candidate for achieving quantum advantage in combinatorial optimization. However, its variational framework presents a long-standing challenge in selecting circuit parameters. In this work, we prove that the energy expectation produced by QAOA can be expressed as a trigonometric function of the final-level mixer parameter. Leveraging this insight, we introduce Penta-O, a level-wise parameter-setting strategy that eliminates the classical outer loop, maintains minimal sampling overhead, and ensures non-decreasing performance. This method is broadly applicable to the generic quadratic unconstrained binary optimization formulated as the Ising model. For a $p$-level QAOA, Penta-O achieves an unprecedented quadratic time complexity of $\mathcal{O}(p^2)$ and a sampling overhead proportional to $5p+1$. Through experiments and simulations, we demonstrate that QAOA enhanced by Penta-O achieves near-optimal performance with exceptional circuit depth efficiency. Our work provides a versatile tool for advancing variational quantum algorithms., Comment: 13 pages, 4 figures

Lower-Limb Fractures (LLF) are a major health concern for older adults, often leading to reduced mobility and prolonged recovery, potentially impairing daily activities and independence. During recovery, older adults frequently face social isolation and functional decline, complicating rehabilitation and adversely affecting physical and mental health. Multi-modal sensor platforms that continuously collect data and analyze it using machine-learning algorithms can remotely monitor this population and infer health outcomes. They can also alert clinicians to individuals at risk of isolation and decline. This paper presents a new publicly available multi-modal sensor dataset, MAISON-LLF, collected from older adults recovering from LLF in community settings. The dataset includes data from smartphone and smartwatch sensors, motion detectors, sleep-tracking mattresses, and clinical questionnaires on isolation and decline. The dataset was collected from ten older adults living alone at home for eight weeks each, totaling 560 days of 24-hour sensor data. For technical validation, supervised machine-learning and deep-learning models were developed using the sensor and clinical questionnaire data, providing a foundational comparison for the research community.

We lay some mathematically rigorous foundations for the resolution of differential equations with respect to semi-classical bases and topologies, namely Freud-Sobolev polynomials and spaces. In this quest, we uncover an elegant theory melding various topics in Numerical and Functional Analysis: Poincar\'e inequalities, Sobolev orthogonal polynomials, Painlev\'e equations and more. Brought together, these ingredients allow us to quantify the compactness of Sobolev embeddings on Freud-weighted spaces and finally resolve some differential equations in this topology. As an application, we rigorously and tightly enclose solutions of the Gross-Pitaevskii equation with sextic potential.

Thin-film solid-state metal dealloying (thin-film SSMD) is a promising method for fabricating nanostructures with controlled morphology and efficiency, offering advantages over conventional bulk materials processing methods for integration into practical applications. Although machine learning (ML) has facilitated the design of dealloying systems, the selection of key thermal treatment parameters for nanostructure formation remains largely unknown and dependent on experimental trial and error. To overcome this challenge, a workflow enabling high-throughput characterization of thermal treatment parameters while probing local nanostructures of thin-film samples is needed. In this work, a laser-based thermal treatment is demonstrated to create temperature gradients on single thin-film samples of Nb-Al/Sc and Nb-Al/Cu. This continuous thermal space enables observation of dealloying transitions and the resulting nanostructures of interest. Through synchrotron X-ray multimodal and high-throughput characterization, critical transitions and nanostructures can be rapidly captured and subsequently verified using electron microscopy. The key temperatures driving chemical reactions and morphological evolutions are clearly identified within this framework. While the oxidation process may contribute to nanostructure formation during thin-film treatment, the dealloying process at the dealloying front involves interactions solely between the dealloying elements, highlighting the availability and viability of the selected systems. This approach enables efficient exploration of the dealloying process and validation of ML predictions, thereby accelerating the discovery of thin-film SSMD systems with targeted nanostructures., Comment: The main content contains 6 figures within 25 pages. The supporting information includes 5 figures within 5 pages

People and businesses increasingly rely on public LLM services, such as ChatGPT, DALLE, and Claude. Understanding their outages, and particularly measuring their failure-recovery processes, is becoming a stringent problem. However, only limited studies exist in this emerging area. Addressing this problem, in this work we conduct an empirical characterization of outages and failure-recovery in public LLM services. We collect and prepare datasets for 8 commonly used LLM services across 3 major LLM providers, including market-leads OpenAI and Anthropic. We conduct a detailed analysis of failure recovery statistical properties, temporal patterns, co-occurrence, and the impact range of outage-causing incidents. We make over 10 observations, among which: (1) Failures in OpenAI's ChatGPT take longer to resolve but occur less frequently than those in Anthropic's Claude;(2) OpenAI and Anthropic service failures exhibit strong weekly and monthly periodicity; and (3) OpenAI services offer better failure-isolation than Anthropic services. Our research explains LLM failure characteristics and thus enables optimization in building and using LLM systems. FAIR data and code are publicly available on https://zenodo.org/records/14018219 and https://github.com/atlarge-research/llm-service-analysis.

Rigged objects are commonly used in artist pipelines, as they can flexibly adapt to different scenes and postures. However, articulating the rigs into realistic affordance-aware postures (e.g., following the context, respecting the physics and the personalities of the object) remains time-consuming and heavily relies on human labor from experienced artists. In this paper, we tackle the novel problem and design A3Syn. With a given context, such as the environment mesh and a text prompt of the desired posture, A3Syn synthesizes articulation parameters for arbitrary and open-domain rigged objects obtained from the Internet. The task is incredibly challenging due to the lack of training data, and we do not make any topological assumptions about the open-domain rigs. We propose using 2D inpainting diffusion model and several control techniques to synthesize in-context affordance information. Then, we develop an efficient bone correspondence alignment using a combination of differentiable rendering and semantic correspondence. A3Syn has stable convergence, completes in minutes, and synthesizes plausible affordance on different combinations of in-the-wild object rigs and scenes., Comment: Project page: https://chuyu.org/research/a3syn

This paper presents results of our system for CoMeDi Shared Task, focusing on Subtask 2: Disagreement Ranking. Our system leverages sentence embeddings generated by the paraphrase-xlm-r-multilingual-v1 model, combined with a deep neural regression model incorporating batch normalization and dropout for improved generalization. By predicting the mean of pairwise judgment differences between annotators, our method explicitly targets disagreement ranking, diverging from traditional "gold label" aggregation approaches. We optimized our system with a customized architecture and training procedure, achieving competitive performance in Spearman correlation against mean disagreement labels. Our results highlight the importance of robust embeddings, effective model architecture, and careful handling of judgment differences for ranking disagreement in multilingual contexts. These findings provide insights into the use of contextualized representations for ordinal judgment tasks and open avenues for further refinement of disagreement prediction models., Comment: Accepted at COMEDI shared Task, Workshop at COLING 2025

Let $p$ be a prime. We prove bounds on short Dirichlet character sums evaluated at a class of homogeneous polynomials in arbitrary dimensions. In every dimension, this bound is nontrivial for sums over boxes with side lengths as short as $p^{1/4 + \kappa}$ for any $\kappa>0$. Our methods capitalize on the relationship between characters mod $p$ and characters over finite field extensions as well as bounds on the multiplicative energy of sets in products of finite fields., Comment: 44 pages

Large Language Models (LLMs) have recently achieved impressive results in complex reasoning tasks through Chain of Thought (CoT) prompting. However, most existing CoT methods rely on using the same prompts, whether manually designed or automatically generated, to handle the entire dataset. This one-size-fits-all approach may fail to meet the specific needs arising from the diversities within a single dataset. To solve this problem, we propose the Clustered Distance-Weighted Chain of Thought (CDW-CoT) method, which dynamically constructs prompts tailored to the characteristics of each data instance by integrating clustering and prompt optimization techniques. Our method employs clustering algorithms to categorize the dataset into distinct groups, from which a candidate pool of prompts is selected to reflect the inherent diversity within the dataset. For each cluster, CDW-CoT trains the optimal prompt probability distribution tailored to their specific characteristics. Finally, it dynamically constructs a unique prompt probability distribution for each test instance, based on its proximity to cluster centers, from which prompts are selected for reasoning. CDW-CoT consistently outperforms traditional CoT methods across six datasets, including commonsense, symbolic, and mathematical reasoning tasks. Specifically, when compared to manual CoT, CDW-CoT achieves an average accuracy improvement of 25.34% on LLaMA2 (13B) and 15.72% on LLaMA3 (8B)., Comment: aaai25(poster)

Modern GPUs, with their specialized hardware like tensor cores, are essential for demanding AI and deep learning applications. This study presents a comprehensive, multi-level microbenchmarking analysis of the NVIDIA Hopper GPU architecture, delving into its performance characteristics and novel features. We benchmark Hopper's memory subsystem latency and throughput, comparing its L2 partitioned cache behavior and global memory access patterns against recent GPU generations, Ampere and Ada Lovelace. Our analysis reveals significant performance differences and architectural improvements in Hopper. A core contribution of this work is a detailed evaluation of Hopper's fourth-generation tensor cores, including their FP8 precision support and the novel asynchronous wgmma instructions, assessing their impact on matrix multiply-accumulate operations. We further investigate the performance implications of other key Hopper innovations: DPX instructions for accelerating dynamic programming algorithms, distributed shared memory (DSM) for inter-SM communication, and the Tensor Memory Accelerator (TMA) for asynchronous data movement. This multi-level approach encompasses instruction-level microbenchmarks, library-level analysis of the Transformer Engine, and application-level benchmarks of tensor core performance within large language models. Our findings provide valuable, in-depth insights for software developers seeking to optimize performance and develop accurate performance models for the Hopper architecture, ultimately contributing to a deeper understanding of its potential for accelerating AI and other computationally intensive workloads., Comment: arXiv admin note: substantial text overlap with arXiv:2402.13499

This work presents the application of reinforcement learning to improve the performance of a highly dynamic hopping system with a parallel mechanism. Unlike serial mechanisms, parallel mechanisms can not be accurately simulated due to the complexity of their kinematic constraints and closed-loop structures. Besides, learning to hop suffers from prolonged aerial phase and the sparse nature of the rewards. To address them, we propose a learning framework to encode long-history feedback to account for the under-actuation brought by the prolonged aerial phase. In the proposed framework, we also introduce a simplified serial configuration for the parallel design to avoid directly simulating parallel structure during the training. A torque-level conversion is designed to deal with the parallel-serial conversion to handle the sim-to-real issue. Simulation and hardware experiments have been conducted to validate this framework.

Recent large language models (LLMs) have tended to leverage sparsity to reduce computations, employing the sparsely activated mixture-of-experts (MoE) technique. MoE introduces four modules, including token routing, token communication, expert computation, and expert parallelism, that impact model quality and training efficiency. To enable versatile usage of MoE models, we introduce FSMoE, a flexible training system optimizing task scheduling with three novel techniques: 1) Unified abstraction and online profiling of MoE modules for task scheduling across various MoE implementations. 2) Co-scheduling intra-node and inter-node communications with computations to minimize communication overheads. 3) To support near-optimal task scheduling, we design an adaptive gradient partitioning method for gradient aggregation and a schedule to adaptively pipeline communications and computations. We conduct extensive experiments with configured MoE layers and real-world MoE models on two GPU clusters. Experimental results show that 1) our FSMoE supports four popular types of MoE routing functions and is more efficient than existing implementations (with up to a 1.42$\times$ speedup), and 2) FSMoE outperforms the state-of-the-art MoE training systems (DeepSpeed-MoE and Tutel) by 1.18$\times$-1.22$\times$ on 1458 MoE layers and 1.19$\times$-3.01$\times$ on real-world MoE models based on GPT-2 and Mixtral using a popular routing function.

A novel structure-preserving numerical method to solve random hyperbolic systems of conservation laws is presented. The method uses a concept of generalized, measure-valued solutions to random conservation laws. This yields a linear partial differential equation with respect to the Young measure and allows to compute the approximation based on linear programming problems. We analyze the structure-preserving properties of the derived numerical method and discuss its advantages and disadvantages. Numerical results for one-dimensional Burgers equation and the isentropic Euler equations and comparisons with stochastic collocation method illustrate the behavior of the proposed numerical method.

Current popular Large Vision-Language Models (LVLMs) are suffering from Hallucinations on Object Attributes (HoOA), leading to incorrect determination of fine-grained attributes in the input images. Leveraging significant advancements in 3D generation from a single image, this paper proposes a novel method to mitigate HoOA in LVLMs. This method utilizes multiview images sampled from generated 3D representations as visual prompts for LVLMs, thereby providing more visual information from other viewpoints. Furthermore, we observe the input order of multiple multiview images significantly affects the performance of LVLMs. Consequently, we have devised Multiview Image Augmented VLM (MIAVLM), incorporating a Multiview Attributes Perceiver (MAP) submodule capable of simultaneously eliminating the influence of input image order and aligning visual information from multiview images with Large Language Models (LLMs). Besides, we designed and employed negative instructions to mitigate LVLMs' bias towards ``Yes" responses. Comprehensive experiments demonstrate the effectiveness of our method., Comment: 2025 IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP 2025)

Dynamic graph augmentation is used to improve the performance of dynamic GNNs. Most methods assume temporal locality, meaning that recent edges are more influential than earlier edges. However, for temporal changes in edges caused by random noise, overemphasizing recent edges while neglecting earlier ones may lead to the model capturing noise. To address this issue, we propose STAA (SpatioTemporal Activity-Aware Random Walk Diffusion). STAA identifies nodes likely to have noisy edges in spatiotemporal dimensions. Spatially, it analyzes critical topological positions through graph wavelet coefficients. Temporally, it analyzes edge evolution through graph wavelet coefficient change rates. Then, random walks are used to reduce the weights of noisy edges, deriving a diffusion matrix containing spatiotemporal information as an augmented adjacency matrix for dynamic GNN learning. Experiments on multiple datasets show that STAA outperforms other dynamic graph augmentation methods in node classification and link prediction tasks., Comment: 2025 IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP 2025)

Using a sample of $(10087\pm44)\times10^{6}$ $J/\psi$ events accumulated with the BESIII detector, we analyze the decays $\eta\rightarrow\pi^+\pi^-l^+l^-$ ($l=e$ or $\mu$) via the process $J/\psi\rightarrow\gamma\eta$. The branching fraction of $\eta\rightarrow\pi^+\pi^-e^+e^-$ is measured to be $\mathcal{B}(\eta\rightarrow\pi^+\pi^-e^+e^-)=(3.07\pm0.12_{\rm{stat.}}\pm0.19_{\rm{syst.}}) \times10^{-4}$. No signal events are observed for the $\eta\rightarrow\pi^{+}\pi^{-}\mu^{+}\mu^{-}$ decay, leading to an upper limit on the branching fraction of $\mathcal{B}(\eta\rightarrow\pi^{+}\pi^{-}\mu^{+}\mu^{-})<4.0\times10^{-7}$ at the 90\% confidence level. Furthermore, the $CP$-violation asymmetry parameter is found to be $\mathcal{A}_{CP}(\eta\rightarrow\pi^{+}\pi^{-}e^{+}e^{-})=(-4.04\pm4.69_{\rm{stat.}}\pm0.14_{\rm{syst.}})\%$, showing no evidence of $CP$-violation with current statistics. Additionally, we extract the transition form factor from the decay amplitude of $\eta\rightarrow\pi^+\pi^-e^+e^-$. Finally, axion-like particles are searched for via the decay $\eta\rightarrow\pi^+\pi^-a, a\rightarrow e^+e^-$, and upper limits on this branching fraction relative to that of $\eta\rightarrow\pi^+\pi^-e^+e^-$ are presented as a function of the axion-like particle mass in the range $5-200\ \mathrm{MeV}/c^{2}$.

We theoretically propose a tunable implementation of symmetry-protected topological phases in a synthetic superlattice, taking advantage of the long coherence time and exquisite spectral resolutions offered by gravity-tilted optical lattice clocks. We describe a protocol similar to Rabi spectroscopy that can be used to probe the distinct topological properties of our system. We then demonstrate how the sensitivity of clocks and interferometers can be improved by the topological robustness to unwanted experimental imperfections. The proposed implementation opens a path to exploit the unique opportunities offered by symmetry-protected topological phases in state-of-the-art quantum sensors., Comment: 13 pages, 7 figures

Based on a data sample of $(10087 \pm 44) \times 10^6$ $J/\psi$ events taken with the BESIII detector, we search for the flavor-changing neutral current charmonium decay $J/\psi \to D^{0} \mu^{+} \mu^{-} + \text{c.c.}$. No significant signal above the background is observed, and the upper limit on its branching fraction is set to be $\mathcal{B}(J/\psi \to D^{0}\mu^{+}\mu^{-} + \text{c.c.} ) < 1.1 \times 10^{-7}$ at the 90% confidence level. This marks the first search for a flavor-changing neutral current charmonium decay involving muons in the final state., Comment: 20 pages, 4 figures

Heated pipe flow is widely used in thermal engineering applications, but the presence of buoyancy force can cause intermittency, or multiple flow states at the same parameter values. Such changes in the flow lead to substantial changes in its heat transfer properties and thereby significant changes in the axial temperature gradient. We therefore introduce a model that features a time-dependent background axial temperature gradient, and consider two temperature boundary conditions -- fixed temperature difference and fixed boundary heat flux. Direct numerical simulations (DNS) are based on the pseudo-spectral framework, and good agreement is achieved between present numerical results and experimental results. The code extends openpipeflow.org and is available at the website. The effect of the axially periodic domain on flow dynamics and heat transfer is examined, using pipes of length L=5D and L=25D. Provided that the flow is fully turbulent, results show close agreement for the mean flow and temperature profiles, and only slight differences in root-mean-square fluctuations. When the flow shows spatial intermittency, heat transfer tends to be overestimated using a short pipe, as shear turbulence fills the domain. This is particularly important when shear turbulence starts to be suppressed at intermediate buoyancy numbers. Finally, at such intermediate buoyancy numbers, we confirm that the decay of localised shear turbulence in the heated pipe flow follows a memoryless process, similar to that in isothermal flow. While isothermal flow then laminarises, convective turbulence in the heated flow can intermittently trigger bursts of shear-like turbulence.

Intra-sentential code-switching (CS) refers to the alternation between languages that happens within a single utterance and is a significant challenge for Automatic Speech Recognition (ASR) systems. For example, when a Vietnamese speaker uses foreign proper names or specialized terms within their speech. ASR systems often struggle to accurately transcribe intra-sentential CS due to their training on monolingual data and the unpredictable nature of CS. This issue is even more pronounced for low-resource languages, where limited data availability hinders the development of robust models. In this study, we propose AdaCS, a normalization model integrates an adaptive bias attention module (BAM) into encoder-decoder network. This novel approach provides a robust solution to CS ASR in unseen domains, thereby significantly enhancing our contribution to the field. By utilizing BAM to both identify and normalize CS phrases, AdaCS enhances its adaptive capabilities with a biased list of words provided during inference. Our method demonstrates impressive performance and the ability to handle unseen CS phrases across various domains. Experiments show that AdaCS outperforms previous state-of-the-art method on Vietnamese CS ASR normalization by considerable WER reduction of 56.2% and 36.8% on the two proposed test sets., Comment: Accepted at ICASSP 2025

Understanding and predicting the diverse conformational states of molecules is crucial for advancing fields such as chemistry, material science, and drug development. Despite significant progress in generative models, accurately generating complex and biologically or material-relevant molecular structures remains a major challenge. In this work, we introduce a diffusion model for three-dimensional (3D) molecule generation that combines a classifiable diffusion model, Diffusion Transformer, with multihead equivariant self-attention. This method addresses two key challenges: correctly attaching hydrogen atoms in generated molecules through learning representations of molecules after hydrogen atoms are removed; and overcoming the limitations of existing models that cannot generate molecules across multiple classes simultaneously. The experimental results demonstrate that our model not only achieves state-of-the-art performance across several key metrics but also exhibits robustness and versatility, making it highly suitable for early-stage large-scale generation processes in molecular design, followed by validation and further screening to obtain molecules with specific properties.

Traditional microperforated panels (MPPs) and metamaterial-based sound absorbers rely on local resonances or multi-resonator designs, which limit their bandwidth, angular applicability, and ease of fabrication. Leveraging the reciprocity theorem and cavity resonances, we introduce a new class of robust MPP absorbers, termed meta-MPPs, capable of achieving ultrabroadband near-total sound absorption across a range of 0.37 to 10 kHz. These absorbers demonstrate average performance exceeding that of traditional MPPs by over 100%, approaching the theoretical causality limit. Notably, their absorption performance can be tuned between angularly asymmetric and omnidirectional modes and remains highly robust to variations in MPP parameters and geometrical configurations. Validated through simulations and experiments, our findings present a simpler, more robust, and highly adaptable solution for noise control.

Flat-band systems provide an ideal platform for exploring exotic quantum phenomena, where the strongly suppressed kinetic energy in these flat energy bands suggests the potential for exotic phases driven by geometric structure, disorder, and interactions. While intriguing phenomena and physical mechanisms have been unveiled in theoretical models, synthesizing such systems within scalable quantum platforms remains challenging. Here, we present the experimental realization of a $\pi$-flux rhombic system using a two-dimensional superconducting qubit array with tunable coupling. We experimentally observe characteristic dynamics, e.g., $\pi$-flux driven destructive interference, and demonstrate the protocol for eigenstate preparation in this rhombic array with coupler-assisted flux. Our results provide future possibilities for exploring the interplay of geometry, interactions, and quantum information encoding in such degenerate systems., Comment: 7+7 pages, 4+2 figures

This work presents a large language model (LLM)-based agent OpenFOAMGPT tailored for OpenFOAM-centric computational fluid dynamics (CFD) simulations, leveraging two foundation models from OpenAI: the GPT-4o and a chain-of-thought (CoT)-enabled o1 preview model. Both agents demonstrate success across multiple tasks. While the price of token with o1 model is six times as that of GPT-4o, it consistently exhibits superior performance in handling complex tasks, from zero-shot case setup to boundary condition modifications, turbulence model adjustments, and code translation. Through an iterative correction loop, the agent efficiently addressed single- and multi-phase flow, heat transfer, RANS, LES, and other engineering scenarios, often converging in a limited number of iterations at low token costs. To embed domain-specific knowledge, we employed a retrieval-augmented generation (RAG) pipeline, demonstrating how preexisting simulation setups can further specialize the agent for sub-domains such as energy and aerospace. Despite the great performance of the agent, human oversight remains crucial for ensuring accuracy and adapting to shifting contexts. Fluctuations in model performance over time suggest the need for monitoring in mission-critical applications. Although our demonstrations focus on OpenFOAM, the adaptable nature of this framework opens the door to developing LLM-driven agents into a wide range of solvers and codes. By streamlining CFD simulations, this approach has the potential to accelerate both fundamental research and industrial engineering advancements.

The search for resonant mass bumps in invariant-mass distributions remains a cornerstone strategy for uncovering Beyond the Standard Model (BSM) physics at the Large Hadron Collider (LHC). Traditional methods often rely on predefined functional forms and exhaustive computational and human resources, limiting the scope of tested final states and selections. This work presents BumpNet, a machine learning-based approach leveraging advanced neural network architectures to generalize and enhance the Data-Directed Paradigm (DDP) for resonance searches. Trained on a diverse dataset of smoothly-falling analytical functions and realistic simulated data, BumpNet efficiently predicts statistical significance distributions across varying histogram configurations, including those derived from LHC-like conditions. The network's performance is validated against idealized likelihood ratio-based tests, showing minimal bias and strong sensitivity in detecting mass bumps across a range of scenarios. Additionally, BumpNet's application to realistic BSM scenarios highlights its capability to identify subtle signals while managing the look-elsewhere effect. These results underscore BumpNet's potential to expand the reach of resonance searches, paving the way for more comprehensive explorations of LHC data in future analyses.

We systematically investigate the observable properties of dark matter-admixed quark stars (DQSs) using the confined-isospin-density-dependent-mass model in combination with the generic bosonic self-interacting dark matter model. Our results show that the dark matter (DM) can significantly influence the properties of quark stars including the mass, radius, and the central pressure at the maximum mass configurations. Moreover, we observe that the mass of DMparticles and the DMfraction significantly affect the types of stellar configurations, and we study these configurations in detail under various scenarios and predict the possibility that two recently observed peculiar objects HESS J1731-347 and PSR J014-4002E are DQSs., Comment: 13 pages, 10 figures

We study the electromagnetic Dalitz decay $J/\psi \to e^+e^- \pi^0$ using $(10087 \pm 44) \times 10^6$ $J/\psi$ events collected by the \bes detector. The di-electron-invariant-mass dependent transition form factor of this decay is explored for the first time. A significant resonant structure corresponding to the $\rho/\omega$ resonance is observed, which cannot be described by existing theoretical models, due to contributions from the isospin-conserving $J/\psi \to \rho \pi^0$ and isospin-volating $J/\psi \to \omega \pi^0$ decays. The observed $\rho$--$\omega$ interference is consistent with that of the pion form factor but features a relatively narrow $\rho$ peak. By taking into account the contribution of this resonant structure, the branching fraction of $J/\psi \to e^+e^- \pi^0$ in the full $e^+e^-$ invariant mass spectrum range is also measured for the first time to be $(8.06 \pm 0.31 (\rm{stat}) \pm 0.38 (\rm{syst}))\times 10^{-7}$, which is two times larger than the prediction of the Vector Meson Dominance model due to the observed resonant contribution of $\rho/\omega$ resonances., Comment: 9 pages, 4 figures, Submitted to Phys. Rev. Lett

Pre-training a language model and then fine-tuning it has shown to be an efficient and effective technique for a wide range of code intelligence tasks, such as code generation, code summarization, and vulnerability detection. However, pretraining language models on a large-scale code corpus is computationally expensive. Fortunately, many off-the-shelf Pre-trained Code Models (PCMs), such as CodeBERT, CodeT5, CodeGen, and Code Llama, have been released publicly. These models acquire general code understanding and generation capability during pretraining, which enhances their performance on downstream code intelligence tasks. With an increasing number of these public pre-trained models, selecting the most suitable one to reuse for a specific task is essential. In this paper, we systematically investigate the reusability of PCMs. We first explore three intuitive model selection methods that select by size, training data, or brute-force fine-tuning. Experimental results show that these straightforward techniques either perform poorly or suffer high costs. Motivated by these findings, we explore learning-based model selection strategies that utilize pre-trained models without altering their parameters. Specifically, we train proxy models to gauge the performance of pre-trained models, and measure the distribution deviation between a model's latent features and the task's labels, using their closeness as an indicator of model transferability. We conduct experiments on 100 widely-used opensource PCMs for code intelligence tasks, with sizes ranging from 42.5 million to 3 billion parameters. The results demonstrate that learning-based selection methods reduce selection time to 100 seconds, compared to 2,700 hours with brute-force fine-tuning, with less than 6% performance degradation across related tasks., Comment: Accepted by IEEE SANER 2025