Koki Hashimoto | Searcharxiv

arXiv2025

oboro: Text-to-Image Synthesis on Limited Data using Flow-based Diffusion Transformer with MMH Attention

This project was conducted as a 2nd-term adopted project of the "Post-5G Information and Communication System Infrastructure Enhancement R&D Project Development of Competitive Generative AI Foundation Models (GENIAC)," a business of the Ministry of Economy, Trade and Industry (METI) and the New Energy and Industrial Technology Development Organization (NEDO). To address challenges such as labor shortages in Japan's anime production industry, this project aims to develop an image generation model from scratch. This report details the technical specifications of the developed image generation model, "oboro:." We have developed "oboro:," a new image generation model built from scratch, using only copyright-cleared images for training. A key characteristic is its architecture, designed to generate high-quality images even from limited datasets. The foundation model weights and inference code are publicly available alongside this report. This project marks the first release of an open-source, commercially-oriented image generation AI fully developed in Japan. AiHUB originated from the OSS community; by maintaining transparency in our development process, we aim to contribute to Japan's AI researcher and engineer community and promote the domestic AI development ecosystem.

cs.AI↗

arXiv2024

Approximation of hyperarithmetic analysis by $\omega$-model reflection

This paper presents two types of results related to hyperarithmetic analysis. First, we introduce new variants of the dependent choice axiom, namely $\mathrm{unique}~\Pi^1_0(\mathrm{resp.}~\Sigma^1_1)\text{-}\mathsf{DC}_0$ and $\mathrm{finite}~\Pi^1_0(\mathrm{resp.}~\Sigma^1_1)\text{-}\mathsf{DC}_0$. These variants imply $\mathsf{ACA}_0^+$ but do not imply $\Sigma^1_1\mathrm{~Induction}$. We also demonstrate that these variants belong to hyperarithmetic analysis and explore their implications with well-known theories in hyperarithmetic analysis. Second, we show that $\mathsf{RFN}^{-1}(\mathsf{ATR}_0)$, a class of theories defined using the $\omega$-model reflection axiom, approximates to some extent hyperarithmetic analysis, and investigate the similarities between this class and hyperarithmetic analysis.

math.LO↗