Qiyao Wang

Qiyao Wang, Beixiong Zheng, Xue Xiong et al.

In the evolving landscape of mobile edge computing (MEC), enhancing communication reliability and computation efficiency to support increasingly stringent low-latency services remains a fundamental challenge. Rotatable antenna (RA) is a promising technology that introduces new spatial degrees of freedom (DoFs) to tackle this challenge. In this letter, we investigate an RA-enabled MEC system where antenna boresight directions can be independently adjusted to proactively improve wireless channel conditions for latency-critical users. We aim to minimize the maximum computation latency by jointly optimizing the MEC server computing resource allocation, receive beamforming, and the deflection angles of all RAs. To address the resulting non-convex problem, we develop an efficient alternating optimization (AO) framework. Specifically, the optimal edge computing resource allocation is derived based on the Karush-Kuhn-Tucker (KKT) conditions. Given the computing resources, the receive beamforming is optimized using semidefinite relaxation (SDR) combined with a bisection search. Furthermore, the RA deflection angles are optimized via fractional programming (FP) and successive convex approximation (SCA). Simulation results verify that the proposed RA-enabled MEC scheme significantly reduces the maximum computation latency compared with conventional benchmark methods.

Qiyao Wang, Xinyi Chen, Longze Chen et al.

Patent examination is a complex, multi-stage process requiring both technical expertise and legal reasoning, increasingly challenged by rising application volumes. Prior benchmarks predominantly view patent examination as discriminative classification or static extraction, failing to capture its inherently interactive and iterative nature, similar to the peer review and rebuttal process in academic publishing. In this paper, we introduce PatRe, the first benchmark that models the full patent examination lifecycle, including Office Action generation and applicant rebuttal. PatRe comprises 480 real-world cases and supports both oracle and retrieval-simulated evaluation settings. Our benchmark reframes patent examination as a dynamic, multi-turn process of justification and response. Extensive experiments across various LLMs reveal critical insights into model performance, including differences between proprietary and open-source models, as well as task asymmetries between examiner analysis and applicant-side rebuttal. These findings highlight both the potential and current limitations of LLMs in modeling complex, real-world legal reasoning and technical novelty judgment in patent examination. We release our code and dataset to facilitate future research on patent examination modeling.

Guhong Chen, Chenghao Sun, Cheng Fu et al.

As code large language models (LLMs) evolve into tool-interactive agents via the Model Context Protocol (MCP), their generalization is increasingly limited by low-quality synthetic data and the diminishing returns of quantity scaling. Moreover, quantity-centric scaling exhibits an early bottleneck that underutilizes trajectory data. We propose TDScaling, a Trajectory Diversity Scaling-based data synthesis framework for code agents that scales performance through diversity rather than raw volume. Under a fixed training budget, increasing trajectory diversity yields larger gains than adding more trajectories, improving the performance-cost trade-off for agent training. TDScaling integrates four innovations: (1) a Business Cluster mechanism that captures real-service logical dependencies; (2) a blueprint-driven multi-agent paradigm that enforces trajectory coherence; (3) an adaptive evolution mechanism that steers synthesis toward long-tail scenarios using Domain Entropy, Reasoning Mode Entropy, and Cumulative Action Complexity to prevent mode collapse; and (4) a sandboxed code tool that mitigates catastrophic forgetting of intrinsic coding capabilities. Experiments on general tool-use benchmarks (BFCL, tau^2-Bench) and code agent tasks (RebenchT, CodeCI, BIRD) demonstrate a win-win outcome: TDScaling improves both tool-use generalization and inherent coding proficiency. We plan to release the full codebase and the synthesized dataset (including 30,000+ tool clusters) upon publication.

Qiyao Wang, Shiwen Ni, Huaren Liu et al.

As the capabilities of Large Language Models (LLMs) continue to advance, the field of patent processing has garnered increased attention within the natural language processing community. However, the majority of research has been concentrated on classification tasks, such as patent categorization and examination, or on short text generation tasks like patent summarization and patent quizzes. In this paper, we introduce a novel and practical task known as Draft2Patent, along with its corresponding D2P benchmark, which challenges LLMs to generate full-length patents averaging 17K tokens based on initial drafts. Patents present a significant challenge to LLMs due to their specialized nature, standardized terminology, and extensive length. We propose a multi-agent framework called AutoPatent which leverages the LLM-based planner agent, writer agents, and examiner agent with PGTree and RRAG to generate lengthy, intricate, and high-quality complete patent documents. The experimental results demonstrate that our AutoPatent framework significantly enhances the ability to generate comprehensive patents across various LLMs. Furthermore, we have discovered that patents generated solely with the AutoPatent framework based on the Qwen2.5-7B model outperform those produced by larger and more powerful LLMs, such as GPT-4o, Qwen2.5-72B, and LLAMA3.1-70B, in both objective metrics and human evaluations. We will make the data and code available upon acceptance at \url{https://github.com/QiYao-Wang/AutoPatent}.

Longze Chen, Lu Wang, Renke Shan et al.

Reward models are crucial for aligning large language models (LLMs) with human values and intentions. Existing approaches follow either Generative (GRMs) or Discriminative (DRMs) paradigms, yet both suffer from limitations: GRMs typically demand costly point-wise supervision, while DRMs produce uncalibrated relative scores that lack probabilistic interpretation. To address these challenges, we introduce a novel reward modeling paradigm: Probabilistic Reward Model (PRM). Instead of modeling reward as a deterministic scalar, our approach treats it as a random variable, learning a full probability distribution for the quality of each response. To make this paradigm practical, we present its closed-form, discrete realization: the Ordinal Probabilistic Reward Model (OPRM), which discretizes the quality score into a finite set of ordinal ratings. Building on OPRM, we propose a data-efficient training strategy called Region Flooding Tuning (RgFT). It enables rewards to better reflect absolute text quality by incorporating quality-level annotations, which guide the model to concentrate the probability mass within corresponding rating sub-regions. Experiments on various reward model benchmarks show that our method improves accuracy by $\textbf{2.9%}\sim\textbf{7.4%}$ compared to prior reward models, demonstrating strong performance and data efficiency. Analysis of the score distribution provides evidence that our method captures not only relative rankings but also absolute quality.

Qiyao Wang, Guhong Chen, Hongbo Wang et al.

Intellectual Property (IP) is a highly specialized domain that integrates technical and legal knowledge, making it inherently complex and knowledge-intensive. Recent advancements in LLMs have demonstrated their potential to handle IP-related tasks, enabling more efficient analysis, understanding, and generation of IP-related content. However, existing datasets and benchmarks focus narrowly on patents or cover limited aspects of the IP field, lacking alignment with real-world scenarios. To bridge this gap, we introduce IPBench, the first comprehensive IP task taxonomy and a large-scale bilingual benchmark encompassing 8 IP mechanisms and 20 distinct tasks, designed to evaluate LLMs in real-world IP scenarios. We benchmark 17 main LLMs, ranging from general purpose to domain-specific, including chat-oriented and reasoning-focused models, under zero-shot, few-shot, and chain-of-thought settings. Our results show that even the top-performing model, DeepSeek-V3, achieves only 75.8% accuracy, indicating significant room for improvement. Notably, open-source IP and law-oriented models lag behind closed-source general-purpose models. To foster future research, we publicly release IPBench, and will expand it with additional tasks to better reflect real-world complexities and support model advancements in the IP domain. We provide the data and code in the supplementary URLs.

Qiyao Wang, Haoran Hu, Longze Chen et al.

With the advancement of multimodal large language models (MLLMs) and coding agents, the website development has shifted from manual programming to agent-based project-level code synthesis. Existing benchmarks rely on idealized assumptions, especially for well-structured, information-rich inputs and static execution settings. In contrast, real-world development is constrained by a critical bottleneck: the semantic misalignment between ambiguous, low-quality instructions from non-expert users and model understanding, which results in a failure mode that we term blind execution. To address this gap, we introduce InteractWeb-Bench, the first multimodal interactive benchmark for website generation under non-expert low-code user conditions. InteractWeb-Bench introduces four types of user agents and persona-driven instruction perturbations to systematically simulate diverse user behaviors, including ambiguity, redundancy, and contradiction, grounded in requirement engineering defect taxonomies. We develop an interactive execution environment for agents, featuring a unified action space comprising Clarify, Implement, Verify, and Submit, enabling iterative intent refinement, code synthesis, and visual feedback-based validation. Extensive experiments and analysis reveal that frontier MLLM-based agents remain trapped in blind execution, exposing limitations in intent recognition and adaptive interaction.

Shiwen Ni, Guhong Chen, Shuaimin Li et al.

In recent years, with the rapid development of the depth and breadth of large language models' capabilities, various corresponding evaluation benchmarks have been emerging in increasing numbers. As a quantitative assessment tool for model performance, benchmarks are not only a core means to measure model capabilities but also a key element in guiding the direction of model development and promoting technological innovation. We systematically review the current status and development of large language model benchmarks for the first time, categorizing 283 representative benchmarks into three categories: general capabilities, domain-specific, and target-specific. General capability benchmarks cover aspects such as core linguistics, knowledge, and reasoning; domain-specific benchmarks focus on fields like natural sciences, humanities and social sciences, and engineering technology; target-specific benchmarks pay attention to risks, reliability, agents, etc. We point out that current benchmarks have problems such as inflated scores caused by data contamination, unfair evaluation due to cultural and linguistic biases, and lack of evaluation on process credibility and dynamic environments, and provide a referable design paradigm for future benchmark innovation.

Qiyao Wang, Hongbo Wang, Longze Chen et al.

Scientific idea generation (SIG) is critical to AI-driven autonomous research, yet existing approaches are often constrained by a static retrieval-then-generation paradigm, leading to homogeneous and insufficiently divergent ideas. In this work, we propose FlowPIE, a tightly coupled retrieval-generation framework that treats literature exploration and idea generation as a co-evolving process. FlowPIE expands literature trajectories via a flow-guided Monte Carlo Tree Search (MCTS) inspired by GFlowNets, using the quality of current ideas assessed by an LLM-based generative reward model (GRM) as a supervised signal to guide adaptive retrieval and construct a diverse, high-quality initial population. Based on this population, FlowPIE models idea generation as a test-time idea evolution process, applying selection, crossover, and mutation with the isolation island paradigm and GRM-based fitness computation to incorporate cross-domain knowledge. It effectively mitigates the information cocoons arising from over-reliance on parametric knowledge and static literature. Extensive evaluations demonstrate that FlowPIE consistently produces ideas with higher novelty, feasibility and diversity compared to strong LLM-based and agent-based frameworks, while enabling reward scaling during test time.

M-A-P Team, Xinrun Du, Yifan Yao et al.

Large language models (LLMs) have demonstrated remarkable proficiency in mainstream academic disciplines such as mathematics, physics, and computer science. However, human knowledge encompasses over 200 specialized disciplines, far exceeding the scope of existing benchmarks. The capabilities of LLMs in many of these specialized fields-particularly in light industry, agriculture, and service-oriented disciplines-remain inadequately evaluated. To address this gap, we present SuperGPQA, a comprehensive benchmark that evaluates graduate-level knowledge and reasoning capabilities across 285 disciplines. Our benchmark employs a novel Human-LLM collaborative filtering mechanism to eliminate trivial or ambiguous questions through iterative refinement based on both LLM responses and expert feedback. Our experimental results reveal significant room for improvement in the performance of current state-of-the-art LLMs across diverse knowledge domains (e.g., the reasoning-focused model DeepSeek-R1 achieved the highest accuracy of 61.82% on SuperGPQA), highlighting the considerable gap between current model capabilities and artificial general intelligence. Additionally, we present comprehensive insights from our management of a large-scale annotation process, involving over 80 expert annotators and an interactive Human-LLM collaborative system, offering valuable methodological guidance for future research initiatives of comparable scope.

Qiyao Wang, Jianguo Huang, Shule Lu et al.

The rapid development of Large Language Models (LLMs) in vertical domains, including intellectual property (IP), lacks a specific evaluation benchmark for assessing their understanding, application, and reasoning abilities. To fill this gap, we introduce IPEval, the first evaluation benchmark tailored for IP agency and consulting tasks. IPEval comprises 2657 multiple-choice questions across four major dimensions: creation, application, protection, and management of IP. These questions span patent rights (inventions, utility models, designs), trademarks, copyrights, trade secrets, and other related laws. Evaluation methods include zero-shot, 5-few-shot, and Chain of Thought (CoT) for seven LLM types, predominantly in English or Chinese. Results show superior English performance by models like GPT series and Qwen series, while Chinese-centric LLMs excel in Chinese tests, albeit specialized IP LLMs lag behind general-purpose ones. Regional and temporal aspects of IP underscore the need for LLMs to grasp legal nuances and evolving laws. IPEval aims to accurately gauge LLM capabilities in IP and spur development of specialized models. Website: \url{https://ipeval.github.io/}

Qiyao Wang, Pengfei Li, Li Zhu et al.

This paper presents our proposed methods to ICDAR 2021 Robust Reading Challenge - Integrated Circuit Text Spotting and Aesthetic Assessment (ICDAR RRC-ICTEXT 2021). For the text spotting task, we detect the characters on integrated circuit and classify them based on yolov5 detection model. We balance the lowercase and non-lowercase by using SynthText, generated data and data sampler. We adopt semi-supervised algorithm and distillation to furtherly improve the model's accuracy. For the aesthetic assessment task, we add a classification branch of 3 classes to differentiate the aesthetic classes of each character. Finally, we make model deployment to accelerate inference speed and reduce memory consumption based on NVIDIA Tensorrt. Our methods achieve 59.1 mAP on task 3.1 with 31 FPS and 306M memory (rank 1), 78.7\% F2 score on task 3.2 with 30 FPS and 306M memory (rank 1).

Qiyao Wang, Ahmed Farahat, Chetan Gupta et al.

Time series data have grown at an explosive rate in numerous domains and have stimulated a surge of time series modeling research. A comprehensive comparison of different time series models, for a considered data analytics task, provides useful guidance on model selection for data analytics practitioners. Data scarcity is a universal issue that occurs in a vast range of data analytics problems, due to the high costs associated with collecting, generating, and labeling data as well as some data quality issues such as missing data. In this paper, we focus on the temporal classification/regression problem that attempts to build a mathematical mapping from multivariate time series inputs to a discrete class label or a real-valued response variable. For this specific problem, we identify two types of scarce data: scarce data with small samples and scarce data with sparsely and irregularly observed time series covariates. Observing that all existing works are incapable of utilizing the sparse time series inputs for proper modeling building, we propose a model called sparse functional multilayer perceptron (SFMLP) for handling the sparsity in the time series covariates. The effectiveness of the proposed SFMLP under each of the two types of data scarcity, in comparison with the conventional deep sequential learning models (e.g., Recurrent Neural Network, and Long Short-Term Memory), is investigated through mathematical arguments and numerical experiments.

Qiyao Wang, Haiyan Wang, Chetan Gupta et al.

In the last few decades, building regression models for non-scalar variables, including time series, text, image, and video, has attracted increasing interests of researchers from the data analytic community. In this paper, we focus on a multivariate time series regression problem. Specifically, we aim to learn mathematical mappings from multiple chronologically measured numerical variables within a certain time interval S to multiple numerical variables of interest over time interval T. Prior arts, including the multivariate regression model, the Seq2Seq model, and the functional linear models, suffer from several limitations. The first two types of models can only handle regularly observed time series. Besides, the conventional multivariate regression models tend to be biased and inefficient, as they are incapable of encoding the temporal dependencies among observations from the same time series. The sequential learning models explicitly use the same set of parameters along time, which has negative impacts on accuracy. The function-on-function linear model in functional data analysis (a branch of statistics) is insufficient to capture complex correlations among the considered time series and suffer from underfitting easily. In this paper, we propose a general functional mapping that embraces the function-on-function linear model as a special case. We then propose a non-linear function-on-function model using the fully connected neural network to learn the mapping from data, which addresses the aforementioned concerns in the existing approaches. For the proposed model, we describe in detail the corresponding numerical implementation procedures. The effectiveness of the proposed model is demonstrated through the application to two real-world problems.

Aniruddha Rajendra Rao, Qiyao Wang, Haiyan Wang et al.

Explosive growth in spatio-temporal data and its wide range of applications have attracted increasing interests of researchers in the statistical and machine learning fields. The spatio-temporal regression problem is of paramount importance from both the methodology development and real-world application perspectives. Given the observed spatially encoded time series covariates and real-valued response data samples, the goal of spatio-temporal regression is to leverage the temporal and spatial dependencies to build a mapping from covariates to response with minimized prediction error. Prior arts, including the convolutional Long Short-Term Memory (CovLSTM) and variations of the functional linear models, cannot learn the spatio-temporal information in a simple and efficient format for proper model building. In this work, we propose two novel extensions of the Functional Neural Network (FNN), a temporal regression model whose effectiveness and superior performance over alternative sequential models have been proven by many researchers. The effectiveness of the proposed spatio-temporal FNNs in handling varying spatial correlations is demonstrated in comprehensive simulation studies. The proposed models are then deployed to solve a practical and challenging precipitation prediction problem in the meteorology field.

Qiyao Wang, Ahmed Farahat, Chetan Gupta et al.

Prognostics is concerned with predicting the future health of the equipment and any potential failures. With the advances in the Internet of Things (IoT), data-driven approaches for prognostics that leverage the power of machine learning models are gaining popularity. One of the most important categories of data-driven approaches relies on a predefined or learned health indicator to characterize the equipment condition up to the present time and make inference on how it is likely to evolve in the future. In these approaches, health indicator forecasting that constructs the health indicator curve over the lifespan using partially observed measurements (i.e., health indicator values within an initial period) plays a key role. Existing health indicator forecasting algorithms, such as the functional Empirical Bayesian approach, the regression-based formulation, a naive scenario matching based on the nearest neighbor, have certain limitations. In this paper, we propose a new `generative + scenario matching' algorithm for health indicator forecasting. The key idea behind the proposed approach is to first non-parametrically fit the underlying health indicator curve with a continuous Gaussian Process using a sample of run-to-failure health indicator curves. The proposed approach then generates a rich set of random curves from the learned distribution, attempting to obtain all possible variations of the target health condition evolution process over the system's lifespan. The health indicator extrapolation for a piece of functioning equipment is inferred as the generated curve that has the highest matching level within the observed period. Our experimental results show the superiority of our algorithm over the other state-of-the-art methods.

Qiyao Wang, Haiyan Wang, Chetan Gupta et al.

Operating envelope is an important concept in industrial operations. Accurate identification for operating envelope can be extremely beneficial to stakeholders as it provides a set of operational parameters that optimizes some key performance indicators (KPI) such as product quality, operational safety, equipment efficiency, environmental impact, etc. Given the importance, data-driven approaches for computing the operating envelope are gaining popularity. These approaches typically use classifiers such as support vector machines, to set the operating envelope by learning the boundary in the operational parameter spaces between the manually assigned `large KPI' and `small KPI' groups. One challenge to these approaches is that the assignment to these groups is often ad-hoc and hence arbitrary. However, a bigger challenge with these approaches is that they don't take into account two key features that are needed to operationalize operating envelopes: (i) interpretability of the envelope by the operator and (ii) implementability of the envelope from a practical standpoint. In this work, we propose a new definition for operating envelope which directly targets the expected magnitude of KPI (i.e., no need to arbitrarily bin the data instances into groups) and accounts for the interpretability and the implementability. We then propose a regularized `GA + penalty' algorithm that outputs an envelope where the user can tradeoff between bias and variance. The validity of our proposed algorithm is demonstrated by two sets of simulation studies and an application to a real-world challenge in the mining processes of a flotation plant.

Qiyao Wang, Shuai Zheng, Ahmed Farahat et al.

Remaining Useful Life (RUL) of an equipment or one of its components is defined as the time left until the equipment or component reaches its end of useful life. Accurate RUL estimation is exceptionally beneficial to Predictive Maintenance, and Prognostics and Health Management (PHM). Data driven approaches which leverage the power of algorithms for RUL estimation using sensor and operational time series data are gaining popularity. Existing algorithms, such as linear regression, Convolutional Neural Network (CNN), Hidden Markov Models (HMMs), and Long Short-Term Memory (LSTM), have their own limitations for the RUL estimation task. In this work, we propose a novel Functional Data Analysis (FDA) method called functional Multilayer Perceptron (functional MLP) for RUL estimation. Functional MLP treats time series data from multiple equipment as a sample of random continuous processes over time. FDA explicitly incorporates both the correlations within the same equipment and the random variations across different equipment's sensor time series into the model. FDA also has the benefit of allowing the relationship between RUL and sensor variables to vary over time. We implement functional MLP on the benchmark NASA C-MAPSS data and evaluate the performance using two popularly-used metrics. Results show the superiority of our algorithm over all the other state-of-the-art methods.