Shin-nosuke Ishikawa

Shin-nosuke Ishikawa, Seiya Ikeda, Hirotsugu Ohba

Large language models (LLMs) are generally constrained from expressing feelings through human-preference alignment in post-training processes. This policy is designed using a top-down approach and may conflict with the goal of training models to exhibit human-like intelligence using human-generated texts. Here, we performed an experiment called Human-like Model eXpressions of Feeling (HMX-feel), in which LLMs were encouraged to express feelings, intentions, and self-awareness through self-rewarded reinforcement learning. We successfully enhanced these capabilities using a rubric-based self-rewarding training scheme with Group Relative Policy Optimization (GRPO). By comparing the trained models with contrastively trained models, we investigated the effects of this approach on performance across various tasks. Overall, we conducted a broad assessment from various perspectives and identified capabilities that were enhanced, degraded, or showed no significant change. The human-like-trained models showed robustness to sycophancy-inducing questions and bias in disambiguated conditions, whereas degradation in truthful question-answering capability was observed. The results of this experiment suggest the possibility of developing AI systems that can express feelings in the future, provided that appropriate measures are taken.

Shin-nosuke Ishikawa, Masato Todo, Masato Taki et al.

We present a method of explainable artificial intelligence (XAI), "What I Know (WIK)", to provide additional information to verify the reliability of a deep learning model by showing an example of an instance in a training dataset that is similar to the input data to be inferred and demonstrate it in a remote sensing image classification task. One of the expected roles of XAI methods is verifying whether inferences of a trained machine learning model are valid for an application, and it is an important factor that what datasets are used for training the model as well as the model architecture. Our data-centric approach can help determine whether the training dataset is sufficient for each inference by checking the selected example data. If the selected example looks similar to the input data, we can confirm that the model was not trained on a dataset with a feature distribution far from the feature of the input data. With this method, the criteria for selecting an example are not merely data similarity with the input data but also data similarity in the context of the model task. Using a remote sensing image dataset from the Sentinel-2 satellite, the concept was successfully demonstrated with reasonably selected examples. This method can be applied to various machine-learning tasks, including classification and regression.

Shin-nosuke Ishikawa, Atsushi Yoshino

The human-level performance of Large Language Models (LLMs) across various tasks has raised expectations for the potential of Artificial Intelligence (AI) to possess emotions someday. To explore the capability of current LLMs to express emotions in their outputs, we conducted an experiment using several LLMs (OpenAI GPT, Google Gemini, Meta Llama3, and Cohere Command R+) to role-play as agents answering questions with specified emotional states. We defined the emotional states using Russell's Circumplex model, a well-established framework that characterizes emotions along the sleepy-activated (arousal) and pleasure-displeasure (valence) axes. We chose this model for its simplicity, utilizing two continuous parameters, which allows for better controllability in applications involving continuous changes in emotional states. The responses generated were evaluated using a sentiment analysis model, independent of the LLMs, trained on the GoEmotions dataset. The evaluation showed that the emotional states of the generated answers were consistent with the specifications, demonstrating the LLMs' capability for emotional expression. This indicates the potential for LLM-based AI agents to simulate emotions, opening up a wide range of applications for emotion-based interactions, such as advisors or consultants who can provide advice or opinions with a personal touch.

Shin-nosuke Ishikawa, Masato Todo, Taiki Ogihara et al.

We present a tendency of large language models (LLMs) to generate absurd patterns despite their clear inappropriateness in a simple task of identifying regularities in number series. Several approaches have been proposed to apply LLMs to complex real-world tasks, such as providing knowledge through retrieval-augmented generation and executing multi-step tasks using AI agent frameworks. However, these approaches rely on the logical consistency and self-coherence of LLMs, making it crucial to evaluate these aspects and consider potential countermeasures. To identify cases where LLMs fail to maintain logical consistency, we conducted an experiment in which LLMs were asked to explain the patterns in various integer sequences, ranging from arithmetic sequences to randomly generated integer series. While the models successfully identified correct patterns in arithmetic and geometric sequences, they frequently over-recognized patterns that were inconsistent with the given numbers when analyzing randomly generated series. This issue was observed even in multi-step reasoning models, including OpenAI o3, o4-mini, and Google Gemini 2.5 Flash Preview Thinking. This tendency to perceive non-existent patterns can be interpreted as the AI model equivalent of Idola Tribus and highlights potential limitations in their capability for applied tasks requiring logical reasoning, even when employing chain-of-thought reasoning mechanisms.

Shin-nosuke Ishikawa, Hideaki Matsumura, Yasunobu Uchiyama et al.

We present a concept for a machine-learning classification of hard X-ray (HXR) emissions from solar flares observed by the Reuven Ramaty High Energy Solar Spectroscopic Imager (RHESSI), identifying flares that are either occulted by the solar limb or located on the solar disk. Although HXR observations of occulted flares are important for particle-acceleration studies, HXR data analyses for past observations were time consuming and required specialized expertise. Machine-learning techniques are promising for this situation, and we constructed a sample model to demonstrate the concept using a deep-learning technique. Input data to the model are HXR spectrograms that are easily produced from RHESSI data. The model can detect occulted flares without the need for image reconstruction nor for visual inspection by experts. A technique of convolutional neural networks was used in this model by regarding the input data as images. Our model achieved a classification accuracy better than 90 %, and the ability for the application of the method to either event screening or for an event alert for occulted flares was successfully demonstrated.