Haihua Chen

Haoxuan Zhang, Ruochi Li, Zhenni Liang et al.

Transparent and standardized documentation is essential for building trustworthy generative AI (GAI) systems. However, existing automated methods for generating model and data cards still face three major challenges: (i) static templates, as most systems rely on fixed query templates that cannot adapt to diverse paper structures or evolving documentation requirements; (ii) information scarcity, since web-scale repositories such as Hugging Face often contain incomplete or inconsistent metadata, leading to missing or noisy information; and (iii) lack of benchmarks, as the absence of standardized datasets and evaluation protocols hinders fair and reproducible assessment of documentation quality. To address these limitations, we propose AdaQE-CG, an Adaptive Query Expansion for Card Generation framework that combines dynamic information extraction with cross-card knowledge transfer. Its Intra-Paper Extraction via Context-Aware Query Expansion (IPE-QE) module iteratively refines extraction queries to recover richer and more complete information from scientific papers and repositories, while its Inter-Card Completion using the MetaGAI Pool (ICC-MP) module fills missing fields by transferring semantically relevant content from similar cards in a curated dataset. In addition, we introduce MetaGAI-Bench, the first large-scale, expert-annotated benchmark for evaluating GAI documentation. Comprehensive experiments across five quality dimensions show that AdaQE-CG substantially outperforms existing approaches, exceeds human-authored data cards, and approaches human-level quality for model cards. Code, prompts, and data are publicly available at: https://github.com/haoxuan-unt2024/AdaQE-CG.

Haoxuan Zhang, Ruochi Li, Yang Zhang et al.

The rapid proliferation of Generative AI necessitates rigorous documentation standards for transparency and governance. However, manual creation of Model and Data Cards is not scalable, while automated approaches lack large-scale, high-fidelity benchmarks for systematic evaluation. We introduce MetaGAI, a comprehensive benchmark comprising 2,541 verified document triplets constructed through semantic triangulation of academic papers, GitHub repositories, and Hugging Face artifacts. Unlike prior single-source datasets, MetaGAI employs a multi-agent framework with specialized Retriever, Generator, and Editor agents, validated through four-dimensional human-in-the-loop assessment, including human evaluation of editor-refined ground truth. We establish a robust evaluation protocol combining automated metrics with validated LLM-as-a-Judge frameworks. Extensive analysis reveals that sparse Mixture-of-Experts architectures achieve superior cost-quality efficiency, while a fundamental trade-off exists between faithfulness and completeness. MetaGAI provides a foundational testbed for benchmarking, training, and analyzing automated Model and Data Card generation methods at scale. Our data and code are available at: https://github.com/haoxuan-unt2024/MetaGAI-Benchmark.

Juhyeon Lee, Wonduk Seo, Junseo Koh et al.

Detecting harmful content in multi turn dialogue requires reasoning over the full conversational context rather than isolated utterances. However, most existing methods rely mainly on models internal parametric knowledge, without explicit grounding in external normative principles. This often leads to inconsistent judgments in socially nuanced contexts, limited interpretability, and redundant reasoning across turns. To address this, we propose RoTRAG, a retrieval augmented framework that incorporates concise human written moral norms, called Rules of Thumb (RoTs), into LLM based harm assessment. For each turn, RoTRAG retrieves relevant RoTs from an external corpus and uses them as explicit normative evidence for turn level reasoning and final severity classification. To improve efficiency, we further introduce a lightweight binary routing classifier that decides whether a new turn requires retrieval grounded reasoning or can reuse existing context. Experiments on ProsocialDialog and Safety Reasoning Multi Turn Dialogue show that RoTRAG consistently improves both harm classification and severity estimation over competitive baselines, with an average relative gain of around 40% in F1 across benchmark datasets and an average relative reduction of 8.4% in distributional error, while reducing redundant computation without sacrificing performance.

Haoxuan Zhang, Ruochi Li, Yang Zhang et al.

Scientific innovation is undergoing a paradigm shift driven by the rapid advancement of Large Language Models (LLMs). As science faces mounting challenges including information overload, disciplinary silos, and diminishing returns on conventional research methods, LLMs are emerging as powerful agents capable not only of enhancing scientific workflows but also of participating in and potentially leading the innovation process. Existing surveys mainly focus on different perspectives, phrases, and tasks in scientific research and discovery, while they have limitations in understanding the transformative potential and role differentiation of LLM. This survey proposes a comprehensive framework to categorize the evolving roles of LLMs in scientific innovation across three hierarchical levels: Evaluator, Collaborator, and Scientist. We distinguish between LLMs' contributions to structured scientific research processes and open-ended scientific discovery, thereby offering a unified taxonomy that clarifies capability boundaries, evaluation criteria, and human-AI interaction patterns at each level. Through an extensive analysis of current methodologies, benchmarks, systems, and evaluation metrics, this survey delivers an in-depth and systematic synthesis on LLM-driven scientific innovation. We present LLMs not only as tools for automating existing processes, but also as catalysts capable of reshaping the epistemological foundations of science itself. This survey offers conceptual clarity, practical guidance, and theoretical foundations for future research, while also highlighting open challenges and ethical considerations in the pursuit of increasingly autonomous AI-driven science. Resources related to this survey can be accessed on GitHub at: https://github.com/haoxuan-unt2024/llm4innovation.

Rafid Ishrak Jahan, Heng Fan, Haihua Chen et al.

Emojis have become ubiquitous in online communication, serving as a universal medium to convey emotions and decorative elements. Their widespread use transcends language and cultural barriers, enhancing understanding and fostering more inclusive interactions. While existing work gained valuable insight into emojis understanding, exploring emojis' capability to serve as a universal sentiment indicator leveraging large language models (LLMs) has not been thoroughly examined. Our study aims to investigate the capacity of emojis to serve as reliable sentiment markers through LLMs across languages and cultures. We leveraged the multimodal capabilities of ChatGPT to explore the sentiments of various representations of emojis and evaluated how well emoji-conveyed sentiment aligned with text sentiment on a multi-lingual dataset collected from 32 countries. Our analysis reveals that the accuracy of LLM-based emoji-conveyed sentiment is 81.43%, underscoring emojis' significant potential to serve as a universal sentiment marker. We also found a consistent trend that the accuracy of sentiment conveyed by emojis increased as the number of emojis grew in text. The results reinforce the potential of emojis to serve as global sentiment indicators, offering insight into fields such as cross-lingual and cross-cultural sentiment analysis on social media platforms. Code: https://github.com/ResponsibleAILab/emoji-universal-sentiment.

Haoxuan Zhang, Ruochi Li, Sarthak Shrestha et al.

Peer review serves as the gatekeeper of science, yet the surge in submissions and widespread adoption of large language models (LLMs) in scholarly evaluation present unprecedented challenges. Recent work has focused on using LLMs to improve review efficiency or generate insightful review content. However, unchecked deficient reviews from both human experts and AI systems threaten to systematically undermine the peer review ecosystem and compromise academic integrity. To address this critical issue, we introduce ReviewGuard, an automated system for detecting and categorizing deficient reviews. ReviewGuard employs a comprehensive four-stage LLM-driven framework that: (1) collects ICLR and NeurIPS papers with their corresponding reviews from OpenReview; (2) annotates review types using GPT-4.1 with human validation; (3) addresses class imbalance and data scarcity through LLM-driven synthetic data augmentation, producing a final corpus of 6,634 papers, 24,657 real reviews, and 46,438 synthetic reviews; and (4) fine-tunes both encoder-based models and open source LLMs. We perform comprehensive feature analysis of the structure and quality of the review text. Compared to sufficient reviews, deficient reviews demonstrate lower rating scores, higher self-reported confidence, reduced structural complexity, and a higher proportion of negative sentiment. AI-generated text detection reveals that, since ChatGPT's emergence, AI-generated reviews have increased dramatically. In the evaluation of deficient review detection models, mixed training with synthetic and real review data provides substantial enhancements to recall and F1 scores on the binary task. This study presents the first LLM-driven system for detecting deficient peer reviews, providing evidence to inform AI governance in peer review while offering valuable insights into human-AI collaboration to maintain academic integrity.

Ruochi Li, Haoxuan Zhang, Edward Gehringer et al.

The surge in scientific submissions has placed increasing strain on the traditional peer-review process, prompting the exploration of large language models (LLMs) for automated review generation. While LLMs demonstrate competence in producing structured and coherent feedback, their capacity for critical reasoning, contextual grounding, and quality sensitivity remains limited. To systematically evaluate these aspects, we propose a comprehensive evaluation framework that integrates semantic similarity analysis and structured knowledge graph metrics to assess LLM-generated reviews against human-written counterparts. We construct a large-scale benchmark of 1,683 papers and 6,495 expert reviews from ICLR and NeurIPS in multiple years, and generate reviews using five LLMs. Our findings show that LLMs perform well in descriptive and affirmational content, capturing the main contributions and methodologies of the original work, with GPT-4o highlighted as an illustrative example, generating 15.74% more entities than human reviewers in the strengths section of good papers in ICLR 2025. However, they consistently underperform in identifying weaknesses, raising substantive questions, and adjusting feedback based on paper quality. GPT-4o produces 59.42% fewer entities than real reviewers in the weaknesses and increases node count by only 5.7% from good to weak papers, compared to 50% in human reviews. Similar trends are observed across all conferences, years, and models, providing empirical foundations for understanding the merits and defects of LLM-generated reviews and informing the development of future LLM-assisted reviewing tools. Data, code, and more detailed results are publicly available at https://github.com/RichardLRC/Peer-Review.

Komala Subramanyam Cherukuri, Pranav Abishai Moses, Aisa Sakata et al.

Oral histories are vital records of lived experience, particularly within communities affected by systemic injustice and historical erasure. Effective and efficient analysis of their oral history archives can promote access and understanding of the oral histories. However, Large-scale analysis of these archives remains limited due to their unstructured format, emotional complexity, and high annotation costs. This paper presents a scalable framework to automate semantic and sentiment annotation for Japanese American Incarceration Oral History. Using LLMs, we construct a high-quality dataset, evaluate multiple models, and test prompt engineering strategies in historically sensitive contexts. Our multiphase approach combines expert annotation, prompt design, and LLM evaluation with ChatGPT, Llama, and Qwen. We labeled 558 sentences from 15 narrators for sentiment and semantic classification, then evaluated zero-shot, few-shot, and RAG strategies. For semantic classification, ChatGPT achieved the highest F1 score (88.71%), followed by Llama (84.99%) and Qwen (83.72%). For sentiment analysis, Llama slightly outperformed Qwen (82.66%) and ChatGPT (82.29%), with all models showing comparable results. The best prompt configurations were used to annotate 92,191 sentences from 1,002 interviews in the JAIOH collection. Our findings show that LLMs can effectively perform semantic and sentiment annotation across large oral history collections when guided by well-designed prompts. This study provides a reusable annotation pipeline and practical guidance for applying LLMs in culturally sensitive archival analysis. By bridging archival ethics with scalable NLP techniques, this work lays the groundwork for responsible use of artificial intelligence in digital humanities and preservation of collective memory. GitHub: https://github.com/kc6699c/LLM4OralHistoryAnalysis.

Yuhan Zhou, Mehri Sattari, Haihua Chen et al.

Next-generation autonomous vehicles (AVs) rely on large volumes of multisource and multimodal ($M^2$) data to support real-time decision-making. In practice, data quality (DQ) varies across sources and modalities due to environmental conditions and sensor limitations, yet AV research has largely prioritized algorithm design over DQ analysis. This work focuses on redundancy as a fundamental but underexplored DQ issue in AV datasets. Using the nuScenes and Argoverse 2 (AV2) datasets, we model and measure redundancy in multisource camera data and multimodal image-LiDAR data, and evaluate how removing redundant labels affects the YOLOv8 object detection task. Experimental results show that selectively removing redundant multisource image object labels from cameras with shared fields of view improves detection. In nuScenes, mAP${50}$ gains from $0.66$ to $0.70$, $0.64$ to $0.67$, and from $0.53$ to $0.55$, on three representative overlap regions, while detection on other overlapping camera pairs remains at the baseline even under stronger pruning. In AV2, $4.1$-$8.6\%$ of labels are removed, and mAP${50}$ stays near the $0.64$ baseline. Multimodal analysis also reveals substantial redundancy between image and LiDAR data. These findings demonstrate that redundancy is a measurable and actionable DQ factor with direct implications for AV performance. This work highlights the role of redundancy as a data quality factor in AV perception and motivates a data-centric perspective for evaluating and improving AV datasets. Code, data, and implementation details are publicly available at: https://github.com/yhZHOU515/RedundancyAD

Yuhan Zhou, Fengjiao Tu, Kewei Sha et al.

Machine learning (ML) technologies have become substantial in practically all aspects of our society, and data quality (DQ) is critical for the performance, fairness, robustness, safety, and scalability of ML models. With the large and complex data in data-centric AI, traditional methods like exploratory data analysis (EDA) and cross-validation (CV) face challenges, highlighting the importance of mastering DQ tools. In this survey, we review 17 DQ evaluation and improvement tools in the last 5 years. By introducing the DQ dimensions, metrics, and main functions embedded in these tools, we compare their strengths and limitations and propose a roadmap for developing open-source DQ tools for ML. Based on the discussions on the challenges and emerging trends, we further highlight the potential applications of large language models (LLMs) and generative AI in DQ evaluation and improvement for ML. We believe this comprehensive survey can enhance understanding of DQ in ML and could drive progress in data-centric AI. A complete list of the literature investigated in this survey is available on GitHub at: https://github.com/haihua0913/awesome-dq4ml.

Changjiang Jiang, Fengchang Yu, Haihua Chen et al.

Complex reasoning over tabular data is crucial in real-world data analysis, yet large language models (LLMs) often underperform due to complex queries, noisy data, and limited numerical capabilities. To address these issues, we propose TabDSR, a framework consisting of: (1) a query decomposer that breaks down complex questions, (2) a table sanitizer that cleans and filters noisy tables, and (3) a program-of-thoughts (PoT)-based reasoner that generates executable code to derive the final answer from the sanitized table. To ensure unbiased evaluation and mitigate data leakage, we introduce a new dataset, CalTab151, specifically designed for complex numerical reasoning over tables. Experimental results demonstrate that TabDSR consistently outperforms existing methods, achieving state-of-the-art (SOTA) performance with 8.79%, 6.08%, and 19.87% accuracy improvement on TAT-QA, TableBench, and TabDSR, respectively. Moreover, our framework integrates seamlessly with mainstream LLMs, providing a robust solution for complex tabular numerical reasoning. These findings highlight the effectiveness of our framework in enhancing LLM performance for complex tabular numerical reasoning. Data and code are available upon request.

Huyen Nguyen, Haoxuan Zhang, Yang Zhang et al.

Reliable evaluation of large language model (LLM)-generated summaries remains an open challenge, particularly across heterogeneous domains and document lengths. We conduct a comprehensive meta-evaluation of 14 automatic summarization metrics and LLM-based evaluators across seven datasets spanning five domains, covering documents from short news articles to long scientific, governmental, and legal texts (2K-27K words) with over 1,500 human-annotated summaries. Our results show that traditional lexical overlap metrics (e.g., ROUGE, BLEU) exhibit weak or negative correlation with human judgments, while task-specific neural metrics and LLM-based evaluators achieve substantially higher alignment, especially for linguistic quality assessment. Leveraging these findings, we propose LLM-ReSum, a self-reflective summarization framework that integrates LLM-based evaluation and generation in a closed feedback loop without model finetuning. Across three domains, LLM-ReSum improves low-quality summaries by up to 33% in factual accuracy and 39% in coverage, with human evaluators preferring refined summaries in 89% of cases. We additionally introduce PatentSumEval, a new human-annotated benchmark for legal document summarization comprising 180 expert-evaluated summaries. All code and datasets will be released in GitHub.

Huyen Nguyen, Haoxuan Zhang, Yang Zhang et al.

Evaluating long document summaries remains the primary bottleneck in summarization research. Existing metrics correlate weakly with human judgments and produce aggregate scores without explaining deficiencies or guiding improvement, preventing effective refinement in applications requiring verifiable accuracy. We introduce LongSumEval, a unified framework bridging evaluation and generation through structured question-answering feedback. The framework operationalizes summary quality as answerability and factual alignment of question-answer pairs, generating interpretable scores and actionable feedback that identifies coverage gaps and factual inconsistencies. This resolves the misalignment where evaluation operates independently of generation objectives. Meta-evaluation of our QA-based evaluation module across seven benchmarks demonstrates substantially stronger agreement with human judgments compared to established metrics. Structured feedback enables significant quality improvements through self-refinement without retraining. By demonstrating that evaluation feedback can serve as executable instructions for generation, this work establishes a generalizable paradigm for aligning assessment with improvement, with direct implications for controllable text generation requiring verifiable accuracy and transparent quality control. All code and datasets will be released in GitHub for reproducibility.

Wonduk Seo, Juhyeon Lee, Junseo Koh et al.

Prompt optimization has emerged as an effective alternative to retraining for improving the performance of Large Language Models (LLMs). However, most existing approaches treat evaluation as a black box, relying solely on numerical scores while offering limited insight into why a prompt succeeds or fails. They also depend heavily on trial-and-error refinements, which are difficult to interpret and control. In this paper, we introduce MA-SAPO, a Multi-Agent framework for Score-Aware Prompt Optimization. Compared to prior methods, MA-SAPO explicitly couples evaluation outcomes with structured reasoning to guide systematic edits. The framework specifically consists of two stages: during the Reasoning Phase, agents collaboratively explain metric scores, diagnose weaknesses, and synthesize targeted refinements that are stored as reusable reasoning assets; during the Test Phase, agents retrieve these assets to analyze optimized prompts and apply only evidence-grounded edits. By turning evaluation signals into interpretable reasoning chains, MA-SAPO produces prompt refinements that are more transparent, auditable, and controllable. Experiments on the HelpSteer1/2 benchmarks demonstrate consistent improvements over single-pass prompting, retrieval-augmented baselines, and prior multi-agent strategies, validating the effectiveness of our approach.

Ngan Tran, Haihua Chen, Ana Cleveland et al.

This study presents a comprehensive bibliometric and topic analysis of the disaster informatics literature published between January 2020 to September 2022. Leveraging a large-scale corpus and advanced techniques such as pre-trained language models and generative AI, we identify the most active countries, institutions, authors, collaboration networks, emergent topics, patterns among the most significant topics, and shifts in research priorities spurred by the COVID-19 pandemic. Our findings highlight (1) countries that were most impacted by the COVID-19 pandemic were also among the most active, with each country having specific research interests, (2) countries and institutions within the same region or share a common language tend to collaborate, (3) top active authors tend to form close partnerships with one or two key partners, (4) authors typically specialized in one or two specific topics, while institutions had more diverse interests across several topics, and (5) the COVID-19 pandemic has influenced research priorities in disaster informatics, placing greater emphasis on public health. We further demonstrate that the field is converging on multidimensional resilience strategies and cross-sectoral data-sharing collaborations or projects, reflecting a heightened awareness of global vulnerability and interdependency. Collecting and quality assurance strategies, data analytic practices, LLM-based topic extraction and summarization approaches, and result visualization tools can be applied to comparable datasets or solve similar analytic problems. By mapping out the trends in disaster informatics, our analysis offers strategic insights for policymakers, practitioners, and scholars aiming to enhance disaster informatics capacities in an increasingly uncertain and complex risk landscape.

Yuhan Zhou, Haihua Chen, Kewei Sha

The next-generation autonomous vehicles (AVs), embedded with frequent real-time decision-making, will rely heavily on a large volume of multisource and multimodal data. In real-world settings, the data quality (DQ) of different sources and modalities usually varies due to unexpected environmental factors or sensor issues. However, both researchers and practitioners in the AV field overwhelmingly concentrate on models/algorithms while undervaluing the DQ. To fulfill the needs of the next-generation AVs with guarantees of functionality, efficiency, and trustworthiness, this paper proposes a novel task-centric and data quality vase framework which consists of five layers: data layer, DQ layer, task layer, application layer, and goal layer. The proposed framework aims to map DQ with task requirements and performance goals. To illustrate, a case study investigating redundancy on the nuScenes dataset proves that partially removing redundancy on multisource image data could improve YOLOv8 object detection task performance. Analysis on multimodal data of image and LiDAR further presents existing redundancy DQ issues. This paper opens up a range of critical but unexplored challenges at the intersection of DQ, task orchestration, and performance-oriented system development in AVs. It is expected to guide the AV community toward building more adaptive, explainable, and resilient AVs that respond intelligently to dynamic environments and heterogeneous data streams. Code, data, and implementation details are publicly available at: https://anonymous.4open.science/r/dq4av-framework/README.md.

Huyen Nguyen, Haihua Chen, Lavanya Pobbathi et al.

Evaluating text summarization has been a challenging task in natural language processing (NLP). Automatic metrics which heavily rely on reference summaries are not suitable in many situations, while human evaluation is time-consuming and labor-intensive. To bridge this gap, this paper proposes a novel method based on large language models (LLMs) for evaluating text summarization. We also conducts a comparative study on eight automatic metrics, human evaluation, and our proposed LLM-based method. Seven different types of state-of-the-art (SOTA) summarization models were evaluated. We perform extensive experiments and analysis on datasets with patent documents. Our results show that LLMs evaluation aligns closely with human evaluation, while widely-used automatic metrics such as ROUGE-2, BERTScore, and SummaC do not and also lack consistency. Based on the empirical comparison, we propose a LLM-powered framework for automatically evaluating and improving text summarization, which is beneficial and could attract wide attention among the community.

Haihua Chen, Ngan Tran, Anand Sagar Thumati et al.

Intrusion detection is an essential task in the cyber threat environment. Machine learning and deep learning techniques have been applied for intrusion detection. However, most of the existing research focuses on the model work but ignores the fact that poor data quality has a direct impact on the performance of a machine learning system. More attention should be paid to the data work when building a machine learning-based intrusion detection system. This article first summarizes existing machine learning-based intrusion detection systems and the datasets used for building these systems. Then the data preparation workflow and quality requirements for intrusion detection are discussed. To figure out how data and models affect machine learning performance, we conducted experiments on 11 HIDS datasets using seven machine learning models and three deep learning models. The experimental results show that BERT and GPT were the best algorithms for HIDS on all of the datasets. However, the performance on different datasets varies, indicating the differences between the data quality of these datasets. We then evaluate the data quality of the 11 datasets based on quality dimensions proposed in this paper to determine the best characteristics that a HIDS dataset should possess in order to yield the best possible result. This research initiates a data quality perspective for researchers and practitioners to improve the performance of machine learning-based intrusion detection.

Haihua Chen, Huyen Nguyen

Citation function and citation sentiment are two essential aspects of citation content analysis (CCA), which are useful for influence analysis, the recommendation of scientific publications. However, existing studies are mostly traditional machine learning methods, although deep learning techniques have also been explored, the improvement of the performance seems not significant due to insufficient training data, which brings difficulties to applications. In this paper, we propose to fine-tune pre-trained contextual embeddings ULMFiT, BERT, and XLNet for the task. Experiments on three public datasets show that our strategy outperforms all the baselines in terms of the F1 score. For citation function identification, the XLNet model achieves 87.2%, 86.90%, and 81.6% on DFKI, UMICH, and TKDE2019 datasets respectively, while it achieves 91.72% and 91.56% on DFKI and UMICH in term of citation sentiment identification. Our method can be used to enhance the influence analysis of scholars and scholarly publications.

Haihua Chen

Purpose: Researchers frequently encounter the following problems when writing scientific articles: (1) Selecting appropriate citations to support the research idea is challenging. (2) The literature review is not conducted extensively, which leads to working on a research problem that others have well addressed. This study focuses on citation recommendation in the related studies section by applying the term function of a citation context, potentially improving the efficiency of writing a literature review. Design/methodology/approach: We present nine term functions with three newly created and six identified from existing literature. Using these term functions as labels, we annotate 531 research papers in three topics to evaluate our proposed recommendation strategy. BM25 and Word2vec with VSM are implemented as the baseline models for the recommendation. Then the term function information is applied to enhance the performance. Findings: The experiments show that the term function-based methods outperform the baseline methods regarding the recall, precision, and F1-score measurement, demonstrating that term functions are useful in identifying valuable citations. Research limitations: The dataset is insufficient due to the complexity of annotating citation functions for paragraphs in the related studies section. More recent deep learning models should be performed to future validate the proposed approach. Practical implications: The citation recommendation strategy can be helpful for valuable citation discovery, semantic scientific retrieval, and automatic literature review generation. Originality/value: The proposed citation function-based citation recommendation can generate intuitive explanations of the results for users, improving the transparency, persuasiveness, and effectiveness of recommender systems.