arXiv subjects

Ming Wang

Publications and source records attributed to Ming Wang.

At least 73 records · Page 4Linked to original sources

arXiv2025

AnnaAgent: Dynamic Evolution Agent System with Multi-Session Memory for Realistic Seeker Simulation

Constrained by the cost and ethical concerns of involving real seekers in AI-driven mental health, researchers develop LLM-based conversational agents (CAs) with tailored configurations, such as profiles, symptoms, and scenarios, to simulate seekers. While these efforts advance AI in mental health, achieving more realistic seeker simulation remains hindered by two key challenges: dynamic evolution and multi-session memory. Seekers' mental states often fluctuate during counseling, which typically spans multiple sessions. To address this, we propose AnnaAgent, an emotional and cognitive dynamic agent system equipped with tertiary memory. AnnaAgent incorporates an emotion modulator and a complaint elicitor trained on real counseling dialogues, enabling dynamic control of the simulator's configurations. Additionally, its tertiary memory mechanism effectively integrates short-term and long-term memory across sessions. Evaluation results, both automated and manual, demonstrate that AnnaAgent achieves more realistic seeker simulation in psychological counseling compared to existing baselines. The ethically reviewed and screened code can be found on https://github.com/sci-m-wang/AnnaAgent.

cs.CL↗

arXiv2025

Observation of ferron transport in ferroelectrics

Ferroelectrics feature spontaneous electric dipolar order reconfigurable via electric fields. Recent theoretical studies of the collective excitations of this electric dipolar order give rise to the hope that "ferron" quasiparticles may complement the magnons of magnetic materials in information and heat management technologies. Yet direct experimental evidence of ferron transport remains elusive. Here we demonstrate efficient ferron injection and detection enabled by ferromagnetic metal contacts, achieving nonlocal signal transmission over micrometer distances in a prototypical ferroelectric PMN-PT. The transmission efficiency can be switched by external magnetic fields that couple to the contacts and gate electric fields that control the ferron excitations. Ferron-based devices open new power saving strategies that employ ferroelectric materials in a future sustainable information society.

physics.app-ph↗

arXiv2025

Benchmarking Post-Training Quantization in LLMs: Comprehensive Taxonomy, Unified Evaluation, and Comparative Analysis

Post-training Quantization (PTQ) technique has been extensively adopted for large language models (LLMs) compression owing to its efficiency and low resource requirement. However, current research lacks a in-depth analysis of the superior and applicable scenarios of each PTQ strategy. In addition, existing algorithms focus primarily on performance, overlooking the trade-off among model size, performance, and quantization bitwidth. To mitigate these confusions, we provide a novel benchmark for LLMs PTQ in this paper. Firstly, in order to support our benchmark, we propose a comprehensive taxonomy for existing mainstream methods by scrutinizing their computational strategies (e.g., optimization-based, compensation-based, etc.). Then, we conduct extensive experiments with the baseline within each class, covering models with various sizes (7B-70B), bitwidths, training levels (LLaMA1/2/3/3.1), architectures (Mixtral, DeepSeekMoE and Mamba) and modality (LLaVA1.5 and VILA1.5) on a wide range of evaluation metrics.Through comparative analysis on the results, we summarize the superior of each PTQ strategy and modelsize-bitwidth trade-off considering the performance. For example, our benchmark reveals that compensation-based technique demonstrates outstanding cross-architecture robustness and extremely low-bit PTQ for ultra large models should be reexamined. Finally, we further accordingly claim that a practical combination of compensation and other PTQ strategy can achieve SOTA various robustness. We believe that our benchmark will provide valuable recommendations for the deployment of LLMs and future research on PTQ approaches.We conduct an repository for our benchmark at https://github.com/zjq0455/PTQ_Benchmark.

cs.LG↗

arXiv2025

Taylor's Theorem and Mean Value Theorem for Random Functions and Random Variables

This study addresses the often-overlooked issue of measurability at intermediate points when applying Taylor's theorems to random functions and random vectors (e.g., likelihood functions with respect to estimators) in statistics. Classical Taylor-related theorems were originally developed for deterministic settings. Consequently, they do not directly extend to stochastic functions and variables and do not inherently guarantee the measurability of intermediate points. In statistical contexts, applying these theorems without properly accounting for randomness can lead to analyses that lack well-defined probabilistic interpretations. Elementary approaches, such as pointwise constructions, are insufficient for handling random quantities and establishing measurable intermediate points. Moreover, some statistical literature has implicitly disregarded this issue, often neglecting the stochastic nature of the problem and assuming that intermediate points are measurable. To address this gap, we develop multivariate Taylor's and mean value theorems tailored for random functions and random variables under mild assumptions. We provide illustrative examples demonstrating the applicability of our results to commonly used statistical methods, including maximum likelihood estimation, $M$-estimation, and profile estimation. Our findings contribute a rigorous foundation for the applications of Taylor expansions in statistics.

stat.OT↗

arXiv2025

MM-InstructEval: Zero-Shot Evaluation of (Multimodal) Large Language Models on Multimodal Reasoning Tasks

The emergence of multimodal large language models (MLLMs) has triggered extensive research in model evaluation. While existing evaluation studies primarily focus on unimodal (vision-only) comprehension and reasoning capabilities, they overlook critical assessments of complex multimodal reasoning tasks that require integrated understanding of both visual and textual contexts. Such multimodal tasks present unique challenges, demanding sophisticated reasoning across multiple modalities and deep comprehension of multimodal contexts. In this paper, we present MM-InstructEval, a comprehensive evaluation framework that incorporates diverse metrics to assess model performance across various multimodal reasoning tasks with vision-text contexts. We conduct extensive zero-shot evaluations on 45 models (including 36 MLLMs) across 16 multimodal datasets, encompassing 6 distinct tasks using 10 different instructions. Our framework introduces multiple innovative metrics, including the 'Best Performance' metric to benchmark peak model capabilities, the 'Mean Relative Gain' metric to assess overall efficacy across models and instructions, the 'Stability' metric to measure robustness, and the 'Adaptability' metric to quantify the compatibility between models and instructions. Through comprehensive evaluation and analysis, we uncover several significant insights about model architectures, instruction formats, and their interactions in multimodal reasoning tasks. Our findings establish new benchmarks for assessing the reasoning capabilities of MLLMs and provide strategic guidance for future developments. To facilitate continued research and evaluation in this field, we release our framework and resources at https://github.com/declare-lab/MM-InstructEval, with an interactive leaderboard available at MM-InstructEval Leaderboard (https://declare-lab.github.io/MM-InstructEval/).

cs.MM↗

arXiv2025

Hypothesis testing under uniform-block covariance structures

A block covariance structure is widely observed across large-scale and high-dimensional datasets in diverse fields such as biology, medicine, engineering, economics, and finance. This pattern entails partitioning a covariance matrix into uniform blocks, where each block exhibits equal variances and covariances. The importance of uniform-block structures lies in their ubiquity, interpretability, and ability to accommodate high dimensionality and data missingness. Despite their prevalence, statistical hypothesis testing under uniform-block covariance structures remains largely unexplored, and unknown statistical properties limit their application in research. To address this gap, we develop a comprehensive framework for joint hypothesis tests of both covariance and mean structures, leveraging a novel block Hadamard product representation of uniform-block matrices. Specifically, we derive closed-form likelihood ratio test statistics and information statistics, explicitly establishing their null distributions. Additionally, we perform simultaneous marginal mean tests under a procedure that controls the false discovery proportion (FDP). Extensive simulations validate the consistency between theoretical and empirical distributions of the joint test statistics, assess the performance of the proposed FDP control procedure, and evaluate the robustness of the joint test statistics against structural disruptions and missing data. Lastly, we apply our methodology to hypothesis testing in a high-dimensional imaging dataset.

stat.ME↗

arXiv2025

DeFine: A Decomposed and Fine-Grained Annotated Dataset for Long-form Article Generation

Long-form article generation (LFAG) presents challenges such as maintaining logical consistency, comprehensive topic coverage, and narrative coherence across extended articles. Existing datasets often lack both the hierarchical structure and fine-grained annotation needed to effectively decompose tasks, resulting in shallow, disorganized article generation. To address these limitations, we introduce DeFine, a Decomposed and Fine-grained annotated dataset for long-form article generation. DeFine is characterized by its hierarchical decomposition strategy and the integration of domain-specific knowledge with multi-level annotations, ensuring granular control and enhanced depth in article generation. To construct the dataset, a multi-agent collaborative pipeline is proposed, which systematically segments the generation process into four parts: Data Miner, Cite Retreiver, Q&A Annotator and Data Cleaner. To validate the effectiveness of DeFine, we designed and tested three LFAG baselines: the web retrieval, the local retrieval, and the grounded reference. We fine-tuned the Qwen2-7b-Instruct model using the DeFine training dataset. The experimental results showed significant improvements in text quality, specifically in topic coverage, depth of information, and content fidelity. Our dataset publicly available to facilitate future research.

cs.CL↗

arXiv2025

Language Models as Continuous Self-Evolving Data Engineers

Large Language Models (LLMs) have demonstrated remarkable capabilities on various tasks, while the further evolvement is limited to the lack of high-quality training data. In addition, traditional training approaches rely too much on expert-labeled data, setting a ceiling on the performance of LLMs. To address this issue, we propose a novel paradigm named LANCE (LANguage models as Continuous self-Evolving data engineers) that enables LLMs to train themselves by autonomously generating, cleaning, reviewing, and annotating data with preference information. Our approach demonstrates that LLMs can serve as continuous self-evolving data engineers, significantly reducing the time and cost of the post-training data construction. Through iterative fine-tuning on Qwen2 series models, we validate the effectiveness of LANCE across various tasks, showing that it can maintain high-quality data generation and continuously improve model performance. Across multiple benchmark dimensions, LANCE results in an average score enhancement of 3.64 for Qwen2-7B and 1.75 for Qwen2-7B-Instruct. This training paradigm with autonomous data construction not only reduces the reliance on human experts or external models but also ensures that the data aligns with human preferences, paving the way for the development of future superintelligent systems that can exceed human capabilities. Codes are available at: https://github.com/Control-derek/LANCE.

cs.CL↗

arXiv2025

Quantitative observability for the Schrödinger equation with an anharmonic oscillator

This paper studies the observability inequalities for the Schrödinger equation associated with an anharmonic oscillator $H=-\frac{\d^2}{\d x^2}+|x|$. We build up the observability inequality over an arbitrarily short time interval $(0,T)$, with an explicit expression for the observation constant $C_{obs}$ in terms of $T$, for some observable set that has a different geometric structure compared to those discussed in \cite{HWW}. We obtain the sufficient conditions and the necessary conditions for observable sets, respectively. We also present counterexamples to demonstrate that half-lines are not observable sets, highlighting a major difference in the geometric properties of observable sets compared to those of Schrödinger operators $H=-\frac{\d^2}{\d x^2}+|x|^{2m}$ with $m\ge 1$. Our approach is based on the following ingredients: First, the use of an Ingham-type spectral inequality constructed in this paper; second, the adaptation of a quantitative unique compactness argument, inspired by the work of Bourgain-Burq-Zworski \cite{Bour13}; third, the application of the Szegö's limit theorem from the theory of Toeplitz matrices, which provides a new mathematical tool for proving counterexamples of observability inequalities.

math.AP↗

arXiv2024

Observability inequality, log-type Hausdorff content and heat equations

This paper studies observability inequalities for heat equations on both bounded domains and the whole space $\mathbb{R}^d$. The observation sets are measured by log-type Hausdorff contents, which are induced by certain log-type gauge functions closely related to the heat kernel. On a bounded domain, we derive the observability inequality for observation sets of positive log-type Hausdorff content. Notably, the aforementioned inequality holds not only for all sets with Hausdorff dimension $s$ for any $s\in (d-1,d]$, but also for certain sets of Hausdorff dimension $d-1$. On the whole space $\mathbb{R}^d$, we establish the observability inequality for observation sets that are thick at the scale of the log-type Hausdorff content. Furthermore, we prove that for the 1-dimensional heat equation on an interval, the Hausdorff content we have chosen is an optimal scale for the observability inequality. To obtain these observability inequalities, we use the adapted Lebeau-Robiano strategy from \cite{Duyckaerts2012resolvent}. For this purpose, we prove the following results at scale of the log-type Hausdorff content, the former being derived from the latter: We establish a spectral inequality/a Logvinenko-Sereda uncertainty principle; we set up a quantitative propagation of smallness of analytic functions; we build up a Remez' inequality; and more fundamentally, we provide an upper bound for the log-type Hausdorff content of a set where a monic polynomial is small, based on an estimate in Lubinsky \cite{Lubinsky1997small}, which is ultimately traced back to the classical Cartan Lemma. In addition, we set up a capacity-based slicing lemma (related to the log-type gauge functions) and establish a quantitative relationship between Hausdorff contents and capacities. These tools are crucial in the studies of the aforementioned propagation of smallness in high-dimensional situations.

math.AP↗

arXiv2024

RoCar: A Relationship Network-based Evaluation Method for Large Language Models

Large language models (LLMs) have received increasing attention. However, due to the complexity of its capabilities, how to rationally evaluate the capabilities of LLMs is still a task to be solved. We propose the RoCar method, which utilizes the defined basic schemas to randomly construct a task graph and generates natural language evaluation tasks based on the task graph to evaluate the reasoning and memory abilities of LLMs respectively. Due to the very large randomness of the task construction process, it is possible to ensure that none of the LLMs to be tested has directly learned the evaluation tasks, guaranteeing the fairness of the evaluation method.

cs.CL↗

arXiv2024

Evidence-based Match-status-Aware Gait Recognition for Out-of-Gallery Gait Identification

Existing gait recognition methods typically identify individuals based on the similarity between probe and gallery samples. However, these methods often neglect the fact that the gallery may not contain identities corresponding to the probes, leading to incorrect recognition.To identify Out-of-Gallery (OOG) gait queries, we propose an Evidence-based Match-status-Aware Gait Recognition (EMA-GR) framework. Inspired by Evidential Deep Learning (EDL), EMA-GR is designed to quantify the uncertainty associated with the match status of recognition. Thus, EMA-GR identifies whether the probe has a counterpart in the gallery. Specifically, we adopt an evidence collector to gather match status evidence from a recognition result pair and parameterize a Dirichlet distribution over the gathered evidence, following the Dempster-Shafer Theory of Evidence (DST). We measure the uncertainty and predict the match status of the recognition results, and thus determine whether the probe is an OOG query.To the best of our knowledge, our method is the first attempt to tackle OOG queries in gait recognition. Moreover, EMA-GR is agnostic against gait recognition methods and improves the robustness against OOG queries. Extensive experiments demonstrate that our method achieves state-of-the-art performance on datasets with OOG queries, and can also generalize well to other identity-retrieval tasks. Importantly, our method surpasses existing state-of-the-art methods by a substantial margin, achieving a 51.26% improvement when the OOG query rate is around 50% on OUMVLP.

cs.CV↗

arXiv2024

Minstrel: Structural Prompt Generation with Multi-Agents Coordination for Non-AI Experts

LLMs have demonstrated commendable performance across diverse domains. Nevertheless, formulating high-quality prompts to assist them in their work poses a challenge for non-AI experts. Existing research in prompt engineering suggests somewhat scattered optimization principles and designs empirically dependent prompt optimizers. Unfortunately, these endeavors lack a structural design, incurring high learning costs and it is not conducive to the iterative updating of prompts, especially for non-AI experts. Inspired by structured reusable programming languages, we propose LangGPT, a structural prompt design framework. Furthermore, we introduce Minstrel, a multi-generative agent system with reflection to automate the generation of structural prompts. Experiments and the case study illustrate that structural prompts generated by Minstrel or written manually significantly enhance the performance of LLMs. Furthermore, we analyze the ease of use of structural prompts through a user survey in our online community.

cs.CL↗

arXiv2024

Log-type ultra-analyticity of elliptic equations with gradient terms

It is well known that every solution of an elliptic equation is analytic if its coefficients are analytic. However, less is known about the ultra-analyticity of such solutions. This work addresses the problem of elliptic equations with lower-order terms, where the coefficients are entire functions of exponential type. We prove that every solution satisfies a quantitative logarithmic ultra-analytic bound and demonstrate that this bound is sharp. The results suggest that the ultra-analyticity of solutions to elliptic equations cannot be expected to achieve the same level of ultra-analyticity as the coefficients.

math.AP↗

arXiv2024

Creating a Microstructure Latent Space with Rich Material Information for Multiphase Alloy Design

The intricate microstructure serves as the cornerstone for the composition/processing-structure-property (CPSP) connection in multiphase alloys. Traditional alloy design methods often overlook microstructural details, which diminishes the reliability and effectiveness of the outcomes. This study introduces an improved alloy design algorithm that integrates authentic microstructural information to establish precise CPSP relationships. The approach utilizes a deep-learning framework based on a variational autoencoder to map real microstructural data to a latent space, enabling the prediction of composition, processing steps, and material properties from the latent space vector. By integrating this deep learning model with a specific sampling strategy in the latent space, a novel, microstructure-centered algorithm for multiphase alloy design is developed. This algorithm is demonstrated through the design of a unified dual-phase steel, and the results are assessed at three performance levels. Moreover, an exploration into the latent vector space of the model highlights its seamless interpolation ability and its rich material information content. Notably, the current configuration of the latent space is particularly advantageous for alloy design, offering an exhaustive representation of microstructure, composition, processing, and property variations essential for multiphase alloys.

cond-mat.mtrl-sci↗

arXiv2024

Hierarchical Retrieval-Augmented Generation Model with Rethink for Multi-hop Question Answering

Multi-hop Question Answering (QA) necessitates complex reasoning by integrating multiple pieces of information to resolve intricate questions. However, existing QA systems encounter challenges such as outdated information, context window length limitations, and an accuracy-quantity trade-off. To address these issues, we propose a novel framework, the Hierarchical Retrieval-Augmented Generation Model with Rethink (HiRAG), comprising Decomposer, Definer, Retriever, Filter, and Summarizer five key modules. We introduce a new hierarchical retrieval strategy that incorporates both sparse retrieval at the document level and dense retrieval at the chunk level, effectively integrating their strengths. Additionally, we propose a single-candidate retrieval method to mitigate the limitations of multi-candidate retrieval. We also construct two new corpora, Indexed Wikicorpus and Profile Wikicorpus, to address the issues of outdated and insufficient knowledge. Our experimental results on four datasets demonstrate that HiRAG outperforms state-of-the-art models across most metrics, and our Indexed Wikicorpus is effective. The code for HiRAG is available at https://github.com/2282588541a/HiRAG

cs.CL↗

arXiv2024

FEEL: A Framework for Evaluating Emotional Support Capability with Large Language Models

Emotional Support Conversation (ESC) is a typical dialogue that can effectively assist the user in mitigating emotional pressures. However, owing to the inherent subjectivity involved in analyzing emotions, current non-artificial methodologies face challenges in effectively appraising the emotional support capability. These metrics exhibit a low correlation with human judgments. Concurrently, manual evaluation methods extremely will cause high costs. To solve these problems, we propose a novel model FEEL (Framework for Evaluating Emotional Support Capability with Large Lan-guage Models), employing Large Language Models (LLMs) as evaluators to assess emotional support capabilities. The model meticulously considers various evaluative aspects of ESC to apply a more comprehensive and accurate evaluation method for ESC. Additionally, it employs a probability distribution approach for a more stable result and integrates an ensemble learning strategy, leveraging multiple LLMs with assigned weights to enhance evaluation accuracy. To appraise the performance of FEEL, we conduct extensive experiments on existing ESC model dialogues. Experimental results demonstrate our model exhibits a substantial enhancement in alignment with human evaluations compared to the baselines. Our source code is available at https://github.com/Ansisy/FEEL.

cs.CL↗

arXiv2024

Interleaved Block-Sparse Transform

Low-complexity Bayes-optimal memory approximate message passing (MAMP) is an efficient signal estimation algorithm in compressed sensing and multicarrier modulation. However, achieving replica Bayes optimality with MAMP necessitates a large-scale right-unitarily invariant transformation, which is prohibitive in practical systems due to its high computational complexity and hardware costs. To solve this difficulty, this letter proposes a low-complexity interleaved block-sparse (IBS) transform, which consists of interleaved multiple low-dimensional transform matrices, aimed at reducing the hardware implementation scale while mitigating performance loss. Furthermore, an IBS cross-domain memory approximate message passing (IBS-CD-MAMP) estimator is developed, comprising a memory linear estimator in the IBS transform domain and a non-linear estimator in the source domain. Numerical results show that the IBS-CD-MAMP offers a reduced implementation scale and lower complexity with excellent performance in IBS-based compressed sensing and interleave frequency division multiplexing systems.

cs.IT↗