arXiv subjects

Qiushi Wang

Publications and source records attributed to Qiushi Wang.

14 recordsLinked to original sources

arXiv2026

Graphitic-C3N4/TiO2(B) S-scheme Heterojunctions for Efficient Photocatalytic H2 Production and Organic Pollution Degradation

Achieving both broad solar-spectrum absorption and strong redox capability is critical for semiconductor photocatalysts in environmental remediation and energy conversion. Herein, an S-scheme heterojunction photocatalyst is constructed by coupling TiO2(B) nanorods with g-C3N4 nanosheets. Its well-matched band structure extends light absorption from the UV to the visible region and enables efficient charge separation. Under simulated sunlight irradiation, the 40 wt% g-C3N4/TiO2(B) heterojunction delivers a H2 evolution rate of 1.98 mmol g-1 h-1 for water reduction with methanol as the sacrificial agent, which is 1.5 and 2.0 times higher than those of pure g-C3N4 and TiO2(B), respectively. When exposed to amoxicillin wastewater instead of methanol solution, the heterojunction degrades 98.2% of amoxicillin and produces 20.70 umol g-1 of H2 within 90 min. Moreover, the heterojunction shows excellent photodegradation activity toward various organic antibiotics and dyes, owing to the S-scheme charge separation mechanism. This work highlights the promising potential of S-scheme heterojunctions for photocatalytic H2 production coupled with organic wastewater treatment.

cond-mat.mtrl-sci↗

arXiv2025

A Scenario-based Stochastic Model of using BESS-based Virtual Transmission Lines in Day-Ahead Unit Commitment

The rapid increase in renewable energy sources (RES) implementation in the power system creates more severe network congestion, which may reduce grid operation efficiency and cause renewable curtailment. Deterministic optimization for the unit commitment shows that battery energy storage system (BESS)-based Virtual Transmission Line (VTL), as an alternative to physical transmission lines, can offer a quick solution for congestion relief, reduced operational costs, and lowered renewable curtailment. This paper aims to evaluate the benefits of VTL when considering Renewable Energy Sources uncertainty. Particularly, this work proposes a scenario-based stochastic security-constrained unit commitment model considering VTL, referred to as SSCUC-VTL. It incorporates the forecast error of RES into the commitment decision for systems with VTL. The performance of applying the VTL strategy is compared to that of adding a new physical transmission line and a standalone BESS. A case study has been conducted on an enhanced IEEE 24-bus test system. The simulation results demonstrate that VTL provides 23% more operational cost reduction than the physical transmission line, and up to 67% more congestion relief than the standalone BESS in a power system with solar and wind generation.

eess.SY↗

arXiv2025

Fine-Tuning LLMs to Analyze Multiple Dimensions of Code Review: A Maximum Entropy Regulated Long Chain-of-Thought Approach

Large Language Models (LLMs) have shown great potential in supporting automated code review due to their impressive capabilities in context understanding and reasoning. However, these capabilities are still limited compared to human-level cognition because they are heavily influenced by the training data. Recent research has demonstrated significantly improved performance through fine-tuning LLMs with code review data. However, compared to human reviewers who often simultaneously analyze multiple dimensions of code review to better identify issues, the full potential of these methods is hampered by the limited or vague information used to fine-tune the models. This paper contributes MelcotCR, a chain-of-thought (COT) fine-tuning approach that trains LLMs with an impressive reasoning ability to analyze multiple dimensions of code review by harnessing long COT techniques to provide rich structured information. To address context loss and reasoning logic loss issues that frequently occur when LLMs process long COT prompts, we propose a solution that combines the Maximum Entropy (ME) modeling principle with pre-defined reasoning pathways in MelcotCR to enable more effective utilization of in-context knowledge within long COT prompts while strengthening the logical tightness of the reasoning process. Empirical evaluations on our curated MelcotCR dataset and the public CodeReviewer dataset reveal that a low-parameter base model, such as 14B Qwen2.5, fine-tuned with MelcotCR can surpass state-of-the-art methods in terms of the accuracy of detecting and describing code issues, with its performance remarkably on par with that of the 671B DeepSeek-R1 model.

cs.SE↗

arXiv2025

FinTeam: A Multi-Agent Collaborative Intelligence System for Comprehensive Financial Scenarios

Financial report generation tasks range from macro- to micro-economics analysis, also requiring extensive data analysis. Existing LLM models are usually fine-tuned on simple QA tasks and cannot comprehensively analyze real financial scenarios. Given the complexity, financial companies often distribute tasks among departments. Inspired by this, we propose FinTeam, a financial multi-agent collaborative system, with a workflow with four LLM agents: document analyzer, analyst, accountant, and consultant. We train these agents with specific financial expertise using constructed datasets. We evaluate FinTeam on comprehensive financial tasks constructed from real online investment forums, including macroeconomic, industry, and company analysis. The human evaluation shows that by combining agents, the financial reports generate from FinTeam achieved a 62.00% acceptance rate, outperforming baseline models like GPT-4o and Xuanyuan. Additionally, FinTeam's agents demonstrate a 7.43% average improvement on FinCUGE and a 2.06% accuracy boost on FinEval. Project is available at https://github.com/FudanDISC/DISC-FinLLM/.

cs.CE↗

arXiv2025

Evaluation of Battery Energy Storage System to Provide Virtual Transmission Service

An immediate need in the transmission system is to find alternative solutions that improve system operation and defer the need for new transmission lines. This study comprehensively evaluates the performance and economic benefits of short-term operation of using battery energy storage systems (BESS) as virtual transmission (VT) to promote power transfer across distant regions. Specifically, this work implements various day-ahead energy scheduling models to analyze the impact of VT on system operation cost, network congestion, model computational time, and market performance. The performance of VT is compared with three alternative network congestion mitigation methods, including building new high-voltage physical transmission lines, cost-driven battery energy storage systems, and network reconfiguration, as well as combinations of two of the aforementioned methods. The benchmark dayahead scheduling model is a traditional security-constrained unit commitment model without system upgrades or other network congestion mitigation. Numerical simulations conducted on the IEEE 24-bus system demonstrate that VT provides a 14% operational cost reduction and 34% congested line relief compared to the base case. Compared to other examined schemes, VT is the only one comparable to physical transmission lines that can provide satisfying congestion relief and operation cost reduction without significantly sacrificing computing time and load payment.

eess.SY↗

arXiv2025

InsQABench: Benchmarking Chinese Insurance Domain Question Answering with Large Language Models

The application of large language models (LLMs) has achieved remarkable success in various fields, but their effectiveness in specialized domains like the Chinese insurance industry remains underexplored. The complexity of insurance knowledge, encompassing specialized terminology and diverse data types, poses significant challenges for both models and users. To address this, we introduce InsQABench, a benchmark dataset for the Chinese insurance sector, structured into three categories: Insurance Commonsense Knowledge, Insurance Structured Database, and Insurance Unstructured Documents, reflecting real-world insurance question-answering tasks.We also propose two methods, SQL-ReAct and RAG-ReAct, to tackle challenges in structured and unstructured data tasks. Evaluations show that while LLMs struggle with domain-specific terminology and nuanced clause texts, fine-tuning on InsQABench significantly improves performance. Our benchmark establishes a solid foundation for advancing LLM applications in the insurance domain, with data and code available at https://github.com/HaileyFamo/InsQABench.git.

cs.CL↗

arXiv2025

Distilling Desired Comments for Enhanced Code Review with Large Language Models

There has been a growing interest in using Large Language Models (LLMs) for code review thanks to their proven proficiency in code comprehension. The primary objective of most review scenarios is to generate desired review comments (DRCs) that explicitly identify issues to trigger code fixes. However, existing LLM-based solutions are not so effective in generating DRCs for various reasons such as hallucination. To enhance their code review ability, they need to be fine-tuned with a customized dataset that is ideally full of DRCs. Nevertheless, such a dataset is not yet available, while manual annotation of DRCs is too laborious to be practical. In this paper, we propose a dataset distillation method, Desiview, which can automatically construct a distilled dataset by identifying DRCs from a code review dataset. Experiments on the CodeReviewer dataset comprising more than 150K review entries show that Desiview achieves an impressive performance of 88.93%, 80.37%, 86.67%, and 84.44% in terms of Precision, Recall, Accuracy, and F1, respectively, surpassing state-of-the-art methods. To validate the effect of such a distilled dataset on enhancing LLMs' code review ability, we first fine-tune the latest LLaMA series (i.e., LLaMA 3 and LLaMA 3.1) to build model Desiview4FT. We then enhance the model training effect through KTO alignment by feeding those review comments identified as non-DRCs to the LLMs, resulting in model Desiview4FA. Verification results indicate that Desiview4FA slightly outperforms Desiview4FT, while both models have significantly improved against the base models in terms of generating DRCs. Human evaluation confirms that both models identify issues more accurately and tend to generate review comments that better describe the issues contained in the code than the base LLMs do.

cs.SE↗

arXiv2024

Preference-Oriented Supervised Fine-Tuning: Favoring Target Model Over Aligned Large Language Models

Alignment, endowing a pre-trained Large language model (LLM) with the ability to follow instructions, is crucial for its real-world applications. Conventional supervised fine-tuning (SFT) methods formalize it as causal language modeling typically with a cross-entropy objective, requiring a large amount of high-quality instruction-response pairs. However, the quality of widely used SFT datasets can not be guaranteed due to the high cost and intensive labor for the creation and maintenance in practice. To overcome the limitations associated with the quality of SFT datasets, we introduce a novel \textbf{p}reference-\textbf{o}riented supervised \textbf{f}ine-\textbf{t}uning approach, namely PoFT. The intuition is to boost SFT by imposing a particular preference: \textit{favoring the target model over aligned LLMs on the same SFT data.} This preference encourages the target model to predict a higher likelihood than that predicted by the aligned LLMs, incorporating assessment information on data quality (i.e., predicted likelihood by the aligned LLMs) into the training process. Extensive experiments are conducted, and the results validate the effectiveness of the proposed method. PoFT achieves stable and consistent improvements over the SFT baselines across different training datasets and base models. Moreover, we prove that PoFT can be integrated with existing SFT data filtering methods to achieve better performance, and further improved by following preference optimization procedures, such as DPO.

cs.CL↗

arXiv2024

BoRA: Bi-dimensional Weight-Decomposed Low-Rank Adaptation

In recent years, Parameter-Efficient Fine-Tuning (PEFT) methods like Low-Rank Adaptation (LoRA) have significantly enhanced the adaptability of large-scale pre-trained models. Weight-Decomposed Low-Rank Adaptation (DoRA) improves upon LoRA by separating the magnitude and direction components of the weight matrix, leading to superior performance. However, DoRA's improvements are limited to the vertical dimension, resulting in an asymmetrical pattern between horizontal and vertical dimensions. This paper introduces BoRA, an innovative extension of LoRA and DoRA, characterized by symmetrical properties across horizontal and vertical dimensions. Our approach optimizes the weight matrix symmetrically by adjusting both column-wise and row-wise magnitudes. Extensive experiments demonstrate that BoRA surpasses state-of-the-art PEFT methods, including LoRA and DoRA, achieving superior results across various benchmarks.

cs.CL↗

arXiv2023

DISC-FinLLM: A Chinese Financial Large Language Model based on Multiple Experts Fine-tuning

We propose Multiple Experts Fine-tuning Framework to build a financial large language model (LLM), DISC-FinLLM. Our methodology improves general LLMs by endowing them with multi-turn question answering abilities, domain text processing capabilities, mathematical computation skills, and retrieval-enhanced generation capabilities. We build a financial instruction-tuning dataset named DISC-FIN-SFT, including instruction samples of four categories (consulting, NLP tasks, computing and retrieval-augmented generation). Evaluations conducted on multiple benchmarks demonstrate that our model performs better than baseline models in various financial scenarios. Further resources can be found at https://github.com/FudanDISC/DISC-FinLLM.

cs.CL↗

arXiv2018

Strong convergence rates of modified truncated EM methods for neutral stochastic differential delay equations

The aim of this paper is to investigate strong convergence of modified truncated Euler-Maruyama method for neutral stochastic differential delay equations introduced in Lan (2018). Strong convergence rates of the given numerical scheme to the exact solutions at fixed time $T$ are obtained under local Lipschitz and Khasminskii-type conditions. Moreover, convergence rates over a time interval $[0,T]$ are also obtained under additional polynomial growth condition on $g$ without the weak monotonicity condition (which is usually the standard assumption to obtain the convergence rate). Two examples are presented to interpret our conclusions.

math.PR↗

arXiv2018

Polynomial stability of exact solution and a numerical method for stochastic differential equations with time-dependent delay

Polynomial stability of exact solution and modified truncated Euler-Maruyama method for stochastic differential equations with time-dependent delay are investigated in this paper. By using the well known discrete semimartingale convergence theorem, sufficient conditions are obtained for both bounded and unbounded delay $δ$ to ensure the polynomial stability of the corresponding numerical approximation. Examples are presented to illustrate the conclusion.

math.PR↗

arXiv2017

A Stability-constrained Optimization Framework for Lur'e Systems with Applications in Power Grids

For many nonlinear control systems, the chosen equilibrium determines both the steady-state efficiency and the dynamic performance. This paper addresses the issue of obtaining an optimal equilibrium in terms of some steady-state operation criteria for a Lur'e-type system and such an equilibrium can also guarantee a sufficiently large stability region in the dynamic domain such that the system can tolerate some given disturbance. For this purpose, a set of computationally tractable algebraic constraints, which can properly represent the stability certificate under the optimization framework, are proposed. The existing methods formulate the dynamic performance under the optimization framework by discretizing the differential-algebraic equations, which are computationally intractable for large-scale Lur'e systems like power grids. Dissimilarly, the introduced approach first constructs the stability region based on quadratic Lyapunov functions. Then, a novel method is proposed to project the stability region onto the feasible domain of the optimization problem such that the stability certificate can be incorporated into the optimization framework easily. In the transient stability-constrained optimal power flow (TSCOPF) problem of power systems, researchers look for a steady-state operating point with the minimum generation costs that can maintain system stability under some given transient disturbances. The proposed approach is applied to develop a scalable TSCOPF framework for power systems. The TSCOPF model is tested on the IEEE 118-Bus power system.

math.OC↗

arXiv2013

The ACS Virgo Cluster Survey. XVII. The Spatial Alignment of Globular Cluster Systems With Early-Type Host Galaxies

We study the azimuthal distribution of globular clusters (GCs) in early-type galaxies and compare them to their host galaxies using data from the ACS Virgo Cluster Survey. We find that in host galaxies with visible elongation (epsilon > 0.2) and intermediate to high luminosities (M_z<-19), the GCs are preferentially aligned along the major axis of the stellar light. The red (metal-rich) GC subpopulations show strong alignment with the major axis of the host galaxy, which supports the notion that these GCs are associated with metal-rich field stars. The metal-rich GCs in lenticular galaxies show signs of being more strongly associated with disks rather than bulges. Surprisingly, we find that the blue (metal-poor) GCs can also show the same correlation. If the metal-poor GCs are part of the early formation of the halo and built up through mergers, then our results support a picture where halo formation and merging occur anisotropically, and where the present day major axis is an indicator of the preferred merging axis.

astro-ph.GA↗