arXiv subjects

Youji Iiguni

Publications and source records attributed to Youji Iiguni.

5 recordsLinked to original sources

arXiv2026

LLMs-Integrated Automatic Hate Speech Recognition Using Controllable Text Generation Models

This paper proposes an automatic speech recognition (ASR) model for hate speech using large language models (LLMs). The proposed method integrates the encoder of the ASR model with the decoder of the LLMs, enabling simultaneous transcription and censorship tasks to prevent the exposure of harmful content. Instruction tuning of the LLM to mask hate-related words with specific tokens requires an annotated hate speech dataset, which is limited. We generate text samples using an LLM with the Chain-of-Thought (CoT) prompting technique guided by cultural context and examples and then convert them into speech samples using a text-to-speech (TTS) system. However, some of them contain non-hate speech samples with hate-related words, which degrades the censorship performance. This paper filters the samples which text classification models correctly label as hate content. By adjusting the threshold for the number of correct answer models, we can control the level of hate in the generated dataset, allowing us to train the LLMs through curriculum learning in a gradual manner. Experimental results show that the proposed method achieves a masking accuracy of 58.6\% for hate-related words, surpassing previous baselines. We also confirm that the curriculum training contributes to the efficiency of both transcription and censorship tasks.

eess.AS↗

arXiv2025

Loss Function Design for Deep Unfolded Sparse Signal Recovery: Supervised and Unsupervised Learning

This paper investigates the impact of loss function design in deep unfolding techniques for sparse signal recovery algorithms. We focus on deep unfolded versions of the fundamental iterative shrinkage thresholding algorithm (ISTA) and the iterative hard thresholding algorithm (IHT). To obtain a guideline for the loss function design, we examine the effect of supervised learning using mean squared error and unsupervised learning using the objective function of the original optimization problem. Our simulation results reveal that the effect of loss function design significantly depends on the convexity of the optimization problem. For convex $\ell_1$-regularized problems, supervised-ISTA achieves better final recovery accuracy but fails to minimize the original objective function, whereas we empirically observe that unsupervised-ISTA converges to a nearly identical solution as conventional ISTA but with accelerated convergence. Conversely, for nonconvex $\ell_0$-regularized problems, both supervised-IHT and unsupervised-IHT converge to better local minima than the original IHT, showing similar performance under the training conditions regardless of the loss function employed. However, when the test conditions differ from the training conditions, our results suggest that unsupervised learning offers better robustness to distribution mismatch. These findings provide valuable insights into the design of effective deep unfolded networks for sparse signal recovery applications.

eess.SP↗

arXiv2025

Optimization-Based Image Restoration under Implementation Constraints in Optical Analog Circuits

Optical analog circuits have attracted attention as promising alternatives to traditional electronic circuits for signal processing tasks due to their potential for low-latency and low-power computations. However, implementing iterative algorithms on such circuits presents challenges, particularly due to the difficulty of performing division operations involving dynamically changing variables and the additive noise introduced by optical amplifiers. In this study, we investigate the feasibility of implementing image restoration algorithms using total variation regularization on optical analog circuits. Specifically, we design the circuit structures for the image restoration with widely used alternating direction method of multipliers (ADMM) and primal dual splitting (PDS). Our design avoids division operations involving dynamic variables and incorporate the impact of additive noise introduced by optical amplifiers. Simulation results show that the effective denoising can be achieved in terms of peak signal to noise ratio (PSNR) and structural similarity index measure (SSIM) even when the circuit noise at the amplifiers is taken into account.

eess.IV↗

arXiv2025

Depth-Aided Color Image Inpainting in Quaternion Domain

In this paper, we propose a depth-aided color image inpainting method in the quaternion domain, called depth-aided low-rank quaternion matrix completion (D-LRQMC). In conventional quaternion-based inpainting techniques, the color image is expressed as a quaternion matrix by using the three imaginary parts as the color channels, whereas the real part is set to zero and has no information. Our approach incorporates depth information as the real part of the quaternion representations, leveraging the correlation between color and depth to improve the result of inpainting. In the proposed method, we first restore the observed image with the conventional LRQMC and estimate the depth of the restored result. We then incorporate the estimated depth into the real part of the observed image and perform LRQMC again. Simulation results demonstrate that the proposed D-LRQMC can improve restoration accuracy and visual quality for various images compared to the conventional LRQMC. These results suggest the effectiveness of the depth information for color image processing in quaternion domain.

eess.IV↗

arXiv2024

Deep Unfolding-Aided Parameter Tuning for Plug-and-Play-Based Video Snapshot Compressive Imaging

Snapshot compressive imaging (SCI) captures high-dimensional data efficiently by compressing it into two-dimensional observations and reconstructing high-dimensional data from two-dimensional observations with various algorithms. The plug-and-play (PnP) method is a promising approach for the video SCI reconstruction because it can leverage both observation models and denoising methods for videos. Since the reconstruction accuracy significantly depends on the choice of noise level parameters, this paper proposes a deep unfolding-based method for tuning these parameters in PnP-based video SCI. For the training of the parameters, we prepare training data from the densely annotated video segmentation dataset, reparametrize the noise level parameters, and apply the checkpointing technique to reduce the required memory. Simulation results show that the trained noise level parameters via the proposed approach exhibit a non-monotonic pattern, which is different from the assumptions in the conventional convergence analyses of PnP-based algorithms. These findings provide new insights into both the application of deep unfolding and the theoretical basis of PnP algorithms.

eess.IV↗