Large Language Models (LLMs) should refuse to answer questions beyond their knowledge. This capability, which we term knowledge-aware refusal, is crucial for factual reliability. However, existing metrics fail to faithfully measure this ability. On the one hand, simple refusal-based metrics are biased by refusal rates and yield inconsistent scores when models exhibit different refusal tendencies. On the other hand, existing calibration metrics are proxy-based, capturing the performance of auxiliary calibration processes rather than the model's actual refusal behavior. In this work, we propose the Refusal Index (RI), a principled metric that measures how accurately LLMs refuse questions they do not know. We define RI as Spearman's rank correlation between refusal probability and error probability. To make RI practically measurable, we design a lightweight two-pass evaluation method that efficiently estimates RI from observed refusal rates across two standard evaluation runs. Extensive experiments across 16 models and 5 datasets demonstrate that RI accurately quantifies a model's intrinsic knowledge-aware refusal capability in factual tasks. Notably, RI remains stable across different refusal rates and provides consistent model rankings independent of a model's overall accuracy and refusal rates. More importantly, RI provides insight into an important but previously overlooked aspect of LLM factuality: while LLMs achieve high accuracy on factual tasks, their refusal behavior can be unreliable and fragile. This finding highlights the need to complement traditional accuracy metrics with the Refusal Index for comprehensive factuality evaluation.

This paper studies an application of a recurrent neural network to clustering method for the S-P chart: a binary data set used widely in education. As the number of students increases, the S-P chart becomes hard to handle. In order to classify the large chart into smaller charts, we present a simple clustering method based on the network dynamics. In the method, the network has multiple fixed points and basins of attraction give clusters corresponding to small S-P charts. In order to evaluate the clustering performance, we present an important feature quantity: average caution index that characterizes singularity of students answer oatterns. Performing fundamental experiments, effectiveness of the method is confirmed.

Generative artificial intelligence (AI) based on large language models (LLMs), such as ChatGPT, has demonstrated remarkable progress across various professional fields, including medicine and education. However, their performance in nutritional education, especially in Japanese national licensure examination for registered dietitians, remains underexplored. This study aimed to evaluate the potential of current LLM-based generative AI models as study aids for nutrition students. Questions from the Japanese national examination for registered dietitians were used as prompts for ChatGPT and three Bing models (Precise, Creative, Balanced), based on GPT-3.5 and GPT-4. Each question was entered into independent sessions, and model responses were analyzed for accuracy, consistency, and response time. Additional prompt engineering, including role assignment, was tested to assess potential performance improvements. Bing-Precise (66.2%) and Bing-Creative (61.4%) surpassed the passing threshold (60%), while Bing-Balanced (43.3%) and ChatGPT (42.8%) did not. Bing-Precise and Bing-Creative generally outperformed others across subject fields except Nutrition Education, where all models underperformed. None of the models consistently provided the same correct responses across repeated attempts, highlighting limitations in answer stability. ChatGPT showed greater consistency in response patterns but lower accuracy. Prompt engineering had minimal effect, except for modest improvement when correct answers and explanations were explicitly provided. While some generative AI models marginally exceeded the passing threshold, overall accuracy and answer consistency remained suboptimal. Moreover, all the models demonstrated notable limitations in answer consistency and robustness. Further advancements are needed to ensure reliable and stable AI-based study aids for dietitian licensure preparation.

The average correct answer rates in the Japanese language and math sections of a national achievement test for students in Japan in fiscal 2025 fell from the previous year, the education ministry said Monday. The correct answer rate for the Japanese language section dropped to 67.0% from 67.8% among elementary school sixth-graders. Among junior high school third-graders, the rate sagged to 54.6% from 58.4%, the lowest level since the current question format was introduced in fiscal 2019. They struggled with writing tasks in particular. For the math section, the rate slid to 58.2% from 63.6% among elementary school sixth-graders and to 48.8% from 53.0% among junior high school third-graders.

Continuous assessment of task difficulty and mental workload is essential in improving the usability and accessibility of interactive systems. Eye tracking data has often been investigated to achieve this ability, with reports on the limited role of standard blink metrics. Here, we propose a new approach to the analysis of eye-blink responses for automated estimation of task difficulty. The core module is a time-frequency representation of eye-blink, which aims to capture the richness of information reflected on blinking. In our first study, we show that this method significantly improves the sensitivity to task difficulty. We then demonstrate how to form a framework where the represented patterns are analyzed with multi-dimensional Long Short-Term Memory recurrent neural networks for their non-linear mapping onto difficulty-related parameters. This framework outperformed other methods that used hand-engineered features. This approach works with any built-in camera, without requiring specialized devices. We conclude by discussing how Rethinking Eye-blink can benefit real-world applications.