arXiv subjects

Haihua Chen

Publications and source records attributed to Haihua Chen.

At least 19 recordsLinked to original sources

arXiv2026

Dissecting Neuro-Symbolic Quality Assurance for Synthetic Oncology Data Generation

Synthetic clinical data generation with large language models addresses the scarcity that limits cancer staging research, but oncology hallucinations are categorically harmful: one clinically impossible staging assignment contaminates every downstream model trained on it. Neuro-symbolic pipelines validate during generation, yet the contribution of individual quality-assurance components remains unclear. We report three controlled studies isolating gate necessity, constraint attribution, and retrieval conditionality, holding generation protocol, diversity thresholds, and fine-tuning hyperparameters constant across adapter conditions. The symbolic gate enforces schema completeness, ontology coverage against the Systematized Nomenclature of Medicine, and staging-logic consistency under American Joint Committee on Cancer eighth-edition rules. Ungated, 29.9% of records contain schema failures and 20.1% contain clinically invalid staging. Schema validation is the load-bearing filter: within the fully gated corpus it rejects 148 of 512 records, ontology grounding a further 24, and staging-logic validation none---the only generator producing logic violations is already excluded on schema, making clinical-logic validation a generator-conditional safeguard rather than the dominant filter. Retrieval augmentation is strongly model-dependent: it improves gate compliance for one generator by 12.5 percentage points, has no measurable effect for a second, and collapses output in a third. Across gated configurations ontology density is largely unchanged, indicating that symbolic validation improves clinical validity rather than vocabulary richness. Symbolic gating therefore buys corpus validity but no commensurate gain on real lung-cancer notes in this study; retrieval should be evaluated per model, and ontology density should not be reported as a proxy for corpus quality.

cs.AI↗

arXiv2026

LAMUS: A Large-Scale Corpus for Legal Argument Mining from U.S. Caselaw using LLMs

Legal argument mining aims to identify and classify the functional components of judicial reasoning, such as facts, issues, rules, analysis, and conclusions. Progress in this area is limited by the lack of large-scale, high-quality annotated datasets for U.S. caselaw, particularly at the state level. This paper introduces LAMUS, a sentence-level legal argument mining corpus constructed from U.S. Supreme Court decisions and Texas criminal appellate opinions. The dataset is created using a data-centric pipeline that combines large-scale case collection, LLM-based automatic annotation, and targeted human-in-the-loop quality refinement. We formulate legal argument mining as a six-class sentence classification task and evaluate multiple general-purpose and legal-domain language models under zero-shot, few-shot, and chain-of-thought prompting strategies, with LegalBERT as a supervised baseline. Results show that chain-of-thought prompting substantially improves LLM performance, while domain-specific models exhibit more stable zero-shot behavior. LLM-assisted verification corrects nearly 20% of annotation errors, improving label consistency. Human verification achieves Cohen's Kappa of 0.85, confirming annotation quality. LAMUS provides a scalable resource and empirical insights for future legal NLP research. All code and datasets can be accessed for reproducibility on GitHub at: https://github.com/LavanyaPobbathi/LAMUS/tree/main

cs.CL↗

arXiv2026

LongSumEval: Question-Answering Based Evaluation and Feedback-Driven Refinement for Long Document Summarization

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.

cs.CL↗

arXiv2026

LLM-ReSum: A Framework for LLM Reflective Summarization through Self-Evaluation

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.

cs.CL↗

arXiv2026

RoTRAG: Rule of Thumb Reasoning for Conversation Harm Detection with Retrieval-Augmented Generation

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.

cs.CL↗

arXiv2026

Evolving Roles of LLMs in Scientific Innovation: Assistant, Collaborator, Scientist, and Evaluator

Large language models (LLMs) are increasingly used in scientific research and discovery, supporting tasks ranging from literature retrieval and synthesis to hypothesis generation, autonomous experimentation, and research evaluation. Existing surveys often conflate scientific research with scientific discovery and typically organize systems by domain, task, or autonomy level alone. In this survey, we propose a four-role framework for understanding LLMs in scientific innovation: Assistant, Collaborator, Scientist, and Evaluator. The framework integrates three complementary dimensions: autonomy level, cognitive function, and scientific innovation, to distinguish research-oriented support from frontier-oriented discovery. We review representative methods, benchmarks, and evaluation practices for each role, examining their capabilities, limitations, and human oversight requirements. Across the literature, Assistant systems are comparatively mature in retrieval and synthesis but remain unreliable in open-ended applications; Collaborator systems expand the space of candidate hypotheses yet struggle with novelty-grounding trade-offs; Scientist systems increasingly automate research workflows but face reliability and safety bottlenecks; and Evaluator systems support review and verification while remaining weak in novelty assessment. We argue that progress in AI for science depends not only on model capability, but also on evaluation, oversight, accountability, and institutional integration.

cs.DL↗

arXiv2026

MetaGAI: A Large-Scale and High-Quality Benchmark for Generative AI Model and Data Card Generation

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.

cs.AI↗

arXiv2026

MA-SAPO: Multi-Agent Reasoning for Score-Aware Prompt Optimization

Prompt optimization has become a practical way to improve the performance of Large Language Models (LLMs) without retraining. However, most existing frameworks treat evaluation as a black box, relying solely on outcome scores without explaining why prompts succeed or fail. Moreover, they involve repetitive trial-and-error refinements that remain implicit, offering limited interpretability or actionable guidance for systematic improvement. In this paper, we propose MA-SAPO: a new Multi-Agent Reasoning for Score Aware Prompt Optimization framework that links evaluation outcomes directly to targeted refinements. Specifically, in the Training Phase, multiple agents interpret evaluation scores, diagnose weaknesses, and generate concrete revision directives, which are stored as reusable reasoning assets. In the Test Phase, an analyzer agent retrieves relevant exemplars and assets for a new prompt, and a refiner agent applies evidence-based edits to improve the prompt and its response. By grounding optimization in structured reasoning, MA-SAPO ensures edits are interpretable, auditable, and controllable. Experiments on the HelpSteer1/2 benchmarks show that our framework consistently outperforms single-pass prompting, retrieval-augmented generation, and prior multi-agent methods across multiple evaluation metrics.

cs.MA↗

arXiv2026

AdaQE-CG: Adaptive Query Expansion for Web-Scale Generative AI Model and Data Card Generation

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.

cs.AI↗

arXiv2026

Modeling and Measuring Redundancy in Multisource Multimodal Data for Autonomous Driving

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

cs.CV↗

arXiv2026

Beyond Human Annotation: Recent Advances in Data Generation Methods for Document Intelligence

The advancement of Document Intelligence (DI) demands large-scale, high-quality training data, yet manual annotation remains a critical bottleneck. While data generation methods are evolving rapidly, existing surveys are constrained by fragmented focuses on single modalities or specific tasks, lacking a unified perspective aligned with real-world workflows. To fill this gap, this survey establishes the first comprehensive technical map for data generation in DI. Data generation is redefined as supervisory signal production, and a novel taxonomy is introduced based on the "availability of data and labels." This framework organizes methodologies into four resource-centric paradigms: Data Augmentation, Data Generation from Scratch, Automated Data Annotation, and Self-Supervised Signal Construction. Furthermore, a multi-level evaluation framework is established to integrate intrinsic quality and extrinsic utility, compiling performance gains across diverse DI benchmarks. Guided by this unified structure, the methodological landscape is dissected to reveal critical challenges such as fidelity gaps and frontiers including co-evolutionary ecosystems. Ultimately, by systematizing this fragmented field, data generation is positioned as the central engine for next-generation DI.

cs.AI↗

arXiv2025

ReviewGuard: Enhancing Deficient Peer Review Detection via LLM-Driven Data Augmentation

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. While recent work has focused on using LLMs to improve review efficiency, unchecked deficient reviews from both human experts and AI systems threaten to systematically undermine academic integrity. To address this issue, we introduce ReviewGuard, an automated system for detecting and categorizing deficient reviews through a four-stage LLM-driven framework: data collection from ICLR and NeurIPS on OpenReview, GPT-4.1 annotation with human validation, synthetic data augmentation yielding 6,634 papers with 24,657 real and 46,438 synthetic reviews, and fine-tuning of encoder-based models and open-source LLMs. Feature analysis reveals that deficient reviews exhibit lower rating scores, higher self-reported confidence, reduced structural complexity, and more negative sentiment than sufficient reviews. AI-generated text detection shows dramatic increases in AI-authored reviews since ChatGPT's emergence. Mixed training with synthetic and real data substantially improves detection performance - for example, Qwen 3-8B achieves recall of 0.6653 and F1 of 0.7073, up from 0.5499 and 0.5606 respectively. This study presents the first LLM-driven system for detecting deficient peer reviews, providing evidence to inform AI governance in peer review. Code, prompts, and data are available at https://github.com/haoxuan-unt2024/ReviewGuard

cs.CL↗

arXiv2025

TabDSR: Decompose, Sanitize, and Reason for Complex Numerical Reasoning in Tabular Data

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.

cs.AI↗

arXiv2025

Unveiling the Merits and Defects of LLMs in Automatic Review Generation for Scientific Papers

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.

cs.CL↗

arXiv2025

Large Language Models for Oral History Understanding with Text Classification and Sentiment Analysis

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.

cs.CL↗

arXiv2025

Exploring Public Perceptions of Generative AI in Libraries: A Social Media Analysis of X Discussions

This study investigates public perceptions of generative artificial intelligence (GenAI) in libraries through a large-scale analysis of posts on X (formerly Twitter). Using a mixed-method approach that combines temporal trend analysis, sentiment classification, and social network analysis, this paper explores how public discourse around GenAI and libraries has evolved over time, the emotional tones that dominate the conversation, and the key users or organizations driving engagement. The findings reveal that discussions are predominantly negative in tone, with surges linked to concerns about ethics and intellectual property. Furthermore, social network analysis identifies both institutional authority and individual bridge users who facilitate cross-domain engagement. The results in this paper contribute to the growing body of literature on GenAI in the library and GLAM (Galleries, Libraries, Archives, and Museums) sectors and offer a real-time, public-facing perspective on the emerging opportunities and concerns GenAI presents.

cs.CY↗

arXiv2025

Disaster Informatics after the COVID-19 Pandemic: Bibliometric and Topic Analysis based on Large-scale Academic Literature

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.

cs.SI↗

arXiv2025

Optimization of Flying Ad Hoc Network Topology and Collaborative Path Planning for Multiple UAVs

Multiple unmanned aerial vehicles (UAVs) play a vital role in monitoring and data collection in wide area environments with harsh conditions. In most scenarios, issues such as real-time data retrieval and real-time UAV positioning are often disregarded, essentially neglecting the communication constraints. In this paper, we comprehensively address both the coverage of the target area and the data transmission capabilities of the flying ad hoc network (FANET). The data throughput of the network is therefore maximized by optimizing the network topology and the UAV trajectories. The resultant optimization problem is effectively solved by the proposed reinforcement learning-based trajectory planning (RL-TP) algorithm and the convex-based topology optimization (C-TOP) algorithm sequentially. The RL-TP optimizes the UAV paths while considering the constraints of FANET. The C-TOP maximizes the data throughput of the network while simultaneously constraining the neighbors and transmit powers of the UAVs, which is shown to be a convex problem that can be efficiently solved in polynomial time. Simulations and field experimental results show that the proposed optimization strategy can effectively plan the UAV trajectories and significantly improve the data throughput of the FANET over the adaptive local minimum spanning tree (A-LMST) and cyclic pruning-assisted power optimization (CPAPO) methods.

cs.MA↗