Accurate citation is critical, yet error rates remain high across scientific literature. We present RefLens, an end-to-end system that automates citation verification from PDF parsing to interactive report generation. Unlike summary- or embedding-based approaches, RefLens performs evidence-grounded verification by extracting verbatim spans from original sources and displaying citation-level cards and a paper-level dashboard. In a 35-participant study, users rated value (M=4.34), trust (M=4.15), and usability (M=4.19) highly, with strong adoption intention (M=4.28).

Multi-hop question answering requires complex reasoning across multiple evidence segments, which often overwhelms retrievalaugmented generation systems with lengthy and noisy contexts, thereby undermining both efficiency and accuracy.While existing prompt compression methods attempt to address this issue, they are typically designed for single-turn queries and fail to capture interdependent reasoning steps.We propose IterCOMP, a unified, training-free prompt compression framework that incorporates multi-hop reasoning within an iterative compression loop.IterCOMP decomposes documents into evidence segments, evaluates question answerability, and generates targeted follow-up questions to iteratively integrate essential evidence, producing a compact, reasoning-oriented prompt.Experiments on MusiQue, 2WikiMultiHopQA, and Hot-potQA demonstrate that IterCOMP achieves substantial improvements in Exact Match and F1 scores while reducing the token budget, outperforming existing baselines and exhibiting robustness as reasoning complexity increases.

The advancement of Large Vision-Language Models (LVLMs) requires precise local region-based reasoning that faithfully grounds the model's logic in actual visual evidence. However, existing datasets face limitations in scalability due to extensive manual annotation and lack of explicit alignment between multi-step reasoning and corresponding image regions, which constrains the evaluation of model trustworthiness. To address these challenges, we propose the Visual Grounding Chain-of-Thought (VG-CoT) dataset, which explicitly links each reasoning step to real visual evidence within the image through a fully automated three-stage pipeline. The pipeline first extracts object- and text-level visual evidence using state-of-the-art detection and OCR models, then generates step-by-step grounded reasoning with GPT-4o, and finally refines the grounding through a rationale-driven open-set detection process. In addition, we introduce a new benchmark that comprehensively evaluates LVLMs reasoning across three complementary dimensions: Rationale Quality, Answer Accuracy, and Reasoning-Answer Alignment. Experiments with representative LVLMs, including LLaVA-1.5 and Qwen2-VL, demonstrate consistent improvements on most evaluation metrics, confirming that VG-CoT effectively enhances trustworthy, evidence-based reasoning while maintaining scalable and cost-efficient dataset construction. The dataset and code will be released publicly upon acceptance to facilitate further research.

Evaluating the multi-hop reasoning capabilities of large language models remains a significant challenge. Although current models achieve strong results on existing multi-hop question answering datasets, such performance often masks two critical vulnerabilities: (1) reliance on internal parametric knowledge rather than adherence to the provided context, and (2) exploitation of dataset shortcuts, such as single-document cues or type-matching, that diminish the need for genuine evidence aggregation across multiple documents. We introduce CRiT-QA (Counterfactual Reasoning with Traps), a dataset explicitly designed to address both limitations. To neutralize reliance on memorized knowledge and enforce strict context dependency, CRiT-QA transforms factual reasoning chains with counterfactual entities. Furthermore, it injects multi-anchor distractor chains, plausible but incorrect reasoning paths that diverge at different hops. These traps require models to follow the entire reasoning process rather than exploiting shallow heuristics. Our experiments show that LLMs exhibit substantial performance degradation on CRiT-QA compared to standard datasets, exposing their vulnerability to counterfactual conditions and distractor traps. CRiT-QA thus serves as a rigorous diagnostic tool for evaluating genuine multi-hop reasoning and provides a foundation for developing more reliable, evidence-grounded LLMs.

This paper incorporates the efficiency of automatic summarization and addresses the challenge of generating personalized summaries tailored to individual users' interests and requirements. To tackle this challenge, we introduce SummPilot, an interaction-based customizable summarization system. SummPilot leverages a large language model to facilitate both automatic and interactive summarization. Users can engage with the system to understand document content and personalize summaries through interactive components such as semantic graphs, entity clustering, and explainable evaluation. Our demo and user studies demonstrate SummPilot's adaptability and usefulness for customizable summarization.

Large Language Models (LLMs) have been increasingly employed for query expansion. However, their generative nature often undermines performance on complex multi-hop retrieval tasks by introducing irrelevant or noisy information. To address this challenge, we propose DeCoR (Decompose and Compress for Retrieval), a framework grounded in structured information refinement. Rather than generating additional content, DeCoR strategically restructures the query's underlying reasoning process and distills supporting evidence from retrieved documents. It consists of two core components tailored to the challenges of multi-hop retrieval: (1) Query Decomposition, which decomposes a complex query into explicit reasoning steps, and (2) Query-aware Document Compression, which synthesizes dispersed evidence from candidate documents into a concise summary relevant to the query. This structured design ensures that the final query representation remains both robust and comprehensive. Experimental results demonstrate that, despite utilizing a relatively small LLM, DeCoR outperforms strong baselines that rely on larger models. This finding underscores that, in complex retrieval scenarios, sophisticatedly leveraging the reasoning and summarization capabilities of LLMs offers a more efficient and effective solution than relying solely on their generative capability.

Deep learning models often learn and exploit spurious correlations in training data, using these non-target features to inform their predictions.Such reliance leads to performance degradation and poor generalization on unseen data.To address these limitations, we introduce a more general form of counterfactual data augmentation, termed counterbias data augmentation, which simultaneously tackles multiple biases (e.g., gender bias, simplicity bias) and enhances out-of-distribution robustness.We present COBA: CounterBias Augmentation, a unified framework that operates at the semantic triple level: first decomposing text into subjectpredicate-object triples, then selectively modifying these triples to disrupt spurious correlations.By reconstructing the text from these adjusted triples, COBA generates counterbias data that mitigates spurious patterns.Through extensive experiments, we demonstrate that COBA not only improves downstream task performance, but also effectively reduces biases and strengthens out-of-distribution resilience, offering a versatile and robust solution to the challenges posed by spurious correlations.

Although pre-trained language models have exhibited great flexibility and versatility with prompt-based few-shot learning, they suffer from the extensive parameter size and limited applicability for inference.Recent studies have suggested that PLMs be used as dataset generators and a tiny task-specific model be trained to achieve efficient inference.However, their applicability to various domains is limited because they tend to generate domain-specific datasets.In this work, we propose a novel approach to universal domain generalization that generates a dataset regardless of the target domain.This allows for generalization of the tiny task model to any domain that shares the label space, thus enhancing the real-world applicability of the dataset generation paradigm.Our experiments indicate that the proposed method accomplishes generalizability across various domains while using a parameter set that is orders of magnitude smaller than PLMs.

The quality of the dataset is crucial for ensuring optimal performance and reliability of downstream task models.However, datasets often contain noisy data inadvertently included during the construction process.Numerous attempts have been made to correct this issue through human annotators.However, hiring and managing human annotators is expensive and time-consuming.As an alternative, recent studies are exploring the use of large language models (LLMs) for data annotation.In this study, we present a case study that extends the application of LLM-based data annotation to enhance the quality of existing datasets through a cleansing strategy.Specifically, we leverage approaches such as chain-of-thought and majority voting to imitate human annotation and classify unrelated documents from the Multi-News dataset, which is widely used for the multi-document summarization task.Through our proposed cleansing method, we introduce an enhanced MULTI-NEWS + .By employing LLMs for data cleansing, we demonstrate an efficient and effective approach to improving dataset quality without relying on expensive human annotation efforts.

In clothing manufacturing, automated equipment systems capable of customized small-scale production using robots are receiving much attention. To flexibly produce various types of clothing using such a system, a monitoring and control environment that is friendly to user decision-making is required. Nevertheless, existing equipment systems do not provide such an environment because they are focused on the operation itself, and there are time and financial limitations in establishing a monitoring and control platform environment in equipment systems intended for small-quantity production. This paper verified data collection, simulation-based analysis, and control on an actual clothing production micro-factory system for verifying the operation behavior. Experimentally, we performed the operation verification of the proposed CPS methodology. We confirmed that control of existing scenarios and simulation-based analysis and control of new scenarios are possible. We expect that the proposed method will provide a user-friendly decision-making environment for existing manufacturing facilities and improve the efficiency of the production line.