The rapid evolution of large language models necessitates effective benchmarks for evaluating their role knowledge, which is essential for establishing connections with the real world and providing more immersive interactions. This paper introduces RoleEval, a bilingual benchmark designed to assess the memorization, utilization, and reasoning capabilities of role knowledge. RoleEval comprises RoleEval-Global (including internationally recognized characters) and RoleEval-Chinese (including characters popular in China), with 6,000 Chinese-English parallel multiple-choice questions focusing on 300 influential people and fictional characters drawn from a variety of domains including celebrities, anime, comics, movies, TV series, games, and fictions. These questions cover basic knowledge and multi-hop reasoning abilities, aiming to systematically probe various aspects such as personal information, relationships, abilities, and experiences of the characters. To maintain high standards, we perform a hybrid quality check process combining both automatic and human verification, ensuring that the questions are diverse, challenging, and discriminative.

No abstract

Question Classification plays an important role in most Question Answering systems. In this paper, we exploit semantic features in Support Vector Machines (SVMs) for Question Classification. We propose a semantic tree kernel to incorporate semantic similarity information. A diverse set of semantic features is evaluated. Experimental results show that SVMs with semantic features, especially semantic classes, can significantly outperform the state-of-the-art systems.

The electrogastrogram (EGG) can detect the gastric electromyogram activity, and then reflect the relative change of the rhythm as well as amplitude of the slow wave of the electromyogram. As EGG has the advantages of convenient, painless, non-invasive and accurate measurement of gastric electromyogram activity, it can not only be used to evaluate the effects of gastromotor drugs and gastrointestinal hormones, but also to distinguish healthy people from functional dyspepsia, patients with gastric cancer and patients with low gastric motility according to the results of parameter analysis in EGG. This paper proposes an EGG signal processing and classification method to realize the potential role of EGG in the diagnosis and management of gastrointestinal diseases. First, EGG signal collection was conducted on normal people and patients, and then the test signal was described as accurately as possible according to some key features of the gastric waveform. Based on the collected data, we developed an indicator that can classify high-dimensional signals and provide an indicator that can distinguish or identify two kinds of signal-related indicators. In this way, EGG signals are associated with specific conditions for clinical diagnosis of gastrointestinal dysrhythmia and even for efficacy evaluation.

Reasonable reduction and controlling of software cost is always a challenge for software companies. To estimate software development cost more precisely, current research effort is focused on improving the measurement of software size or complexity by combining or adjusting key cost drivers, such as function points and other observable project context factors. However, personnel factors are seldom investigated or treated in depth as a way to reduce the estimated software development cost. On the premise that a software project is decomposed in a number of tasks, and that predetermined developers are available as resources for it, this paper intends to optimize the allocation of available personnel for lower development cost. In this research, we consider the problem of allocating competent developers to suitable tasks as an unbalanced personnel assignment problem, and improve the traditional Hungarian Algorithm by applying three strategies to find optimal personnel allocation solutions for diverse requirements. Moreover, the performance of our improved algorithms is evaluated and compared through a series of experiments on simulation datasets to identify and validate the measurement indicators and influence factors of their performance.