AI VTubers, where the performer is not human but algorithmically generated, introduce a new context for fandom. While human VTubers have been substantially studied for their cultural appeal, parasocial dynamics, and community economies, little is known about how audiences engage with their AI counterparts. To address this gap, we present a qualitative study of Neuro-sama, the most prominent AI VTuber. Our findings show that engagement is anchored in active co-creation: audiences are drawn by the AI’s unpredictable yet entertaining interactions, cement loyalty through collective emotional events that trigger anthropomorphic projection, and sustain attachment via the AI’s consistent persona. Financial support emerges not as a reward for performance but as a participatory mechanism for shaping livestream content, establishing a resilient fan economy built on ongoing interaction. These dynamics reveal how AI Vtuber fandom reshapes fan–creator relationships and offer implications for designing transparent and sustainable AI-mediated communities.

Multimodal Large Language Models (MLLMs) have emerged to tackle the challenges of Visual Question Answering (VQA), sparking a new research focus on conducting objective evaluations of these models. Existing evaluation methods face limitations due to the significant human workload required to design Q&A pairs for visual images, which inherently restricts the scale and scope of evaluations. Although automated MLLM-as-judge approaches attempt to reduce the human workload through automatic evaluations, they often introduce biases. To address these problems, we propose an Unsupervised Peer review MLLM Evaluation framework. It utilizes only image data, allowing models to automatically generate questions and conduct peer review assessments of answers from other models, effectively alleviating the reliance on human workload. Additionally, we introduce the vision-language scoring system to mitigate the bias issues, which focuses on three aspects: (i) response correctness; (ii) visual understanding and reasoning; and (iii) image-text correlation. Experimental results demonstrate that UPME achieves a Pearson correlation of 0.944 with human evaluations on the MMstar dataset and 0.814 on the ScienceQA dataset, indicating that our framework closely aligns with human-designed benchmarks and inherent human preferences.

LLM-as-a-Judge has been widely utilized as an evaluation method in various benchmarks and served as supervised rewards in model training. However, despite their excellence in many domains, potential issues are under-explored, undermining their reliability and the scope of their utility. Therefore, we identify 12 key potential biases and propose a new automated bias quantification framework—CALM—which systematically quantifies and analyzes each type of bias in LLM-as-a-Judge by using automated and principle-guided modification. Our experiments cover multiple popular language models, and the results indicate that while advanced models have achieved commendable overall performance, significant biases persist in certain specific tasks. Empirical results suggest that there remains room for improvement in the reliability of LLM-as-a-Judge. Moreover, we also discuss the explicit and implicit influence of these biases and give some suggestions for the reliable application of LLM-as-a-Judge. Our work highlights the need for stakeholders to address these issues and remind users to exercise caution in LLM-as-a-Judge applications.

Chinese calligraphy offers fruitful visual structure not found in abstract/ figurative paintings or photographic images. It makes it well-suited for studying how personality shapes aesthetic preference, yet few works have explored this link. This paper introduces the first computational framework that models the link between viewer personality and Kai Shu calligraphic preference. It collects a dataset of Kai Shu calligraphy images, user preference scores, and Big Five personality traits. It extracts 160 structural feature descriptors from eight categories, such as stroke curvature, layout, and whitespace. Regression and attribution methods reveal five patterns, such as visual structure predicts perceived style, and that traits like Openness and Neuroticism influence preference patterns. High-Openness users prefer balanced, clean layouts. Low-Neuroticism users favor lighter, irregular forms. Some personality-feature pairs follow inverted-U trends, where moderate structural complexity leads to higher preference. These results connect cognitive traits with visual structure and support interpretable, personality aware modeling of aesthetic response. Our findings support personalized style discovery and open up new directions for interest-driven aesthetic education and digital preservation. Code available at https://github.com/tianchengliu18/kai2trait.

Large Language Models (LLMs) often struggle to maintain their original performance when faced with semantically coherent but task-irrelevant contextual information. Although prior studies have explored this issue using fixed-template or retrieval-based distractions, such static methods show limited effectiveness against contemporary models. To address this problem, we propose a dynamic distraction generation framework based on tree search, where the generation process is guided by model behavior. Without modifying the original question or answer, the method efficiently produces challenging adaptive distractions across multiple datasets, enabling systematic stress testing of LLMs’ contextual robustness. Experiments on four benchmarks demonstrate that the generated distractions lead to an average performance drop of over 45\% for mainstream models. Further comparisons of mitigation strategies show that prompt-based optimization methods yield limited gains, whereas post-training approaches (e.g., DPO) significantly enhance the model's contextual robustness. The results indicate that these issues do not stem from knowledge deficits in LLMs, but from a fundamental inability to maintain consistent reasoning under contextual distraction, posing a major challenge to the reliability of LLMs in real-world applications.

A compact cost-effective Narrowband Internet-of-Things (NB-IoT) System-in-Package (SiP) in mass production is implemented in 55nm CMOS and Integrated Passive Device (IPD) process. The RF front-end components, i.e., inductors and capacitors, are integrated into an IPD die to reduce the Bill of Material (BOM) cost. Notably, the design eliminated the necessity of the antenna switches (ANTSW). On the IPD die, a diplexer is employed to separate the low-band (LB) and high-band (HB) paths, while in each band, the transmitter (TX) and the receiver (RX) share a common RF low pass filter through a co-matching network. To ensure robust performance, CMOS switches are placed at the PA’s drain and the LNA’s input, protecting the LNA core devices from breaking down and tuning essential impedance matching for both TX and RX. The built-in self-calibration scheme enables the chip to calibrate its TX/RX gain autonomously and flexibly, eliminating the use of external equipment therefore reducing the cost at the module production line. The NB-IoT SoC is integrated with the IPD die in a$7\times 7$QFN package. The system operates across the NB-IoT bands from 699MHz to 2200MHz (the CMOS chip supports 450MHz to 2200MHz), with transmitter’s Psatover +26dBm and achieving the HD2/HD3 below −36dBm when transmitting at +23dBm. The self-calibration scheme improves TX LO leakage and image by more than 15dB and ensures TX/RX gain accuracy within +/−0.5dB across the band and gain states.

Use-after-free vulnerabilities are gaining more and more attentions in recent years, since they are commonly exploited in applications like browsers, and exposed in abundant security updates, e.g., from Microsoft, Google or Mozilla. This kind of vulnerabilities are triggered by dereferencing a dangling pointer, and may introduce high risks into the system once they are exploited. In this paper, we propose a comprehensive solution called UAFChecker to detect use-after-free vulnerabilities in source code. Our solution utilizes classical static analysis techniques, including taint analysis and symbolic execution, to make an inter-procedural analysis to find as many use-after-free vulnerabilities as possible, with a low false negative rate and a low false positive rate. We implement a prototype of UAFChecker based on the compiler framework LLVM. We then use the Juliet Test Suite to evaluate UAFChecker's capability of detecting use-after-free vulnerabilities. Results show that UAFChecker is able to identify most use-after-free vulnerabilities in the Juliet Test Suite. We also test UAFChecker against two open source applications, and successfully find out all known use-after-free vulnerabilities in them.