Ryotaro Shimizu

Ryotaro Shimizu, Yu Wang, Masanari Kimura et al.

In this work, we propose a fashion item recommendation model that incorporates hyperbolic geometry into user and item representations. Using hyperbolic space, our model aims to capture implicit hierarchies among items based on their visual data and users' purchase history. During training, we apply a multi-task learning framework that considers both hyperbolic and Euclidean distances in the loss function. Our experiments on three data sets show that our model performs better than previous models trained in Euclidean space only, confirming the effectiveness of our model. Our ablation studies show that multi-task learning plays a key role, and removing the Euclidean loss substantially deteriorates the model performance.

Ryotaro Shimizu, Masanari Kimura, Masayuki Goto

Several techniques to map various types of components, such as words, attributes, and images, into the embedded space have been studied. Most of them estimate the embedded representation of target entity as a point in the projective space. Some models, such as Word2Gauss, assume a probability distribution behind the embedded representation, which enables the spread or variance of the meaning of embedded target components to be captured and considered in more detail. We examine the method of estimating embedded representations as probability distributions for the interpretation of fashion-specific abstract and difficult-to-understand terms. Terms, such as "casual," "adult-casual,'' "beauty-casual," and "formal," are extremely subjective and abstract and are difficult for both experts and non-experts to understand, which discourages users from trying new fashion. We propose an end-to-end model called dual Gaussian visual-semantic embedding, which maps images and attributes in the same projective space and enables the interpretation of the meaning of these terms by its broad applications. We demonstrate the effectiveness of the proposed method through multifaceted experiments involving image and attribute mapping, image retrieval and re-ordering techniques, and a detailed theoretical/analytical discussion of the distance measure included in the loss function.

Ryotaro Shimizu, Takashi Wada, Yu Wang et al.

Recent research on explainable recommendation generally frames the task as a standard text generation problem, and evaluates models simply based on the textual similarity between the predicted and ground-truth explanations. However, this approach fails to consider one crucial aspect of the systems: whether their outputs accurately reflect the users' (post-purchase) sentiments, i.e., whether and why they would like and/or dislike the recommended items. To shed light on this issue, we introduce new datasets and evaluation methods that focus on the users' sentiments. Specifically, we construct the datasets by explicitly extracting users' positive and negative opinions from their post-purchase reviews using an LLM, and propose to evaluate systems based on whether the generated explanations 1) align well with the users' sentiments, and 2) accurately identify both positive and negative opinions of users on the target items. We benchmark several recent models on our datasets and demonstrate that achieving strong performance on existing metrics does not ensure that the generated explanations align well with the users' sentiments. Lastly, we find that existing models can provide more sentiment-aware explanations when the users' (predicted) ratings for the target items are directly fed into the models as input. The datasets and benchmark implementation are available at: https://github.com/jchanxtarov/sent_xrec.

Yuki Hirakawa, Takashi Wada, Kazuya Morishita et al.

Fashion aesthetic evaluation is the task of estimating how well the outfits worn by individuals in images suit them. In this work, we examine the zero-shot performance of GPT-4V on this task for the first time. We show that its predictions align fairly well with human judgments on our datasets, and also find that it struggles with ranking outfits in similar colors. The code is available at https://github.com/st-tech/gpt4v-fashion-aesthetic-evaluation.

Tianwei Chen, Takuya Furusawa, Yuki Hirakawa et al.

This paper introduces a multi-label visual emotion analysis benchmark dataset for comprehensively evaluating the ability of multimodal large language models (MLLMs) to predict the emotions evoked by images. Recent user studies report an unintuitive finding: humans may prefer the predictions of MLLMs over the labels in existing datasets. We argue that this phenomenon stems from the suboptimal annotation scheme used in existing datasets, where each annotator is shown a single candidate emotion for each image and judges whether it is evoked or not. This approach is clearly limited because a single image can evoke multiple emotions with varying intensities. As a result, evaluations based on these datasets may underestimate the capabilities of MLLMs, yet an appropriate benchmark for evaluating such models remains lacking. To address this issue, we introduce a new multi-label benchmark dataset for visual emotion analysis toward MLLMs evaluation. We hire $20$ annotators per image and ask them to select all emotions they feel from an image. Then, we aggregate the votes across all annotators, providing a more reliable and representative dataset labeled with a distribution of emotions. The resulting dataset contains $10,344$ images with $236,998$ valid votes across eight emotions. Based on this benchmark dataset, we evaluate several recent models, including Qwen3-VL, OpenAI's GPT, Gemini, and Claude. We assess model performance on both dominant emotion prediction and emotion distribution prediction. Our results demonstrate the progress achieved by recent MLLMs while also indicating that substantial room for improvement remains. Furthermore, our experiments with LLM-as-a-judge show that the method does not consistently improve MLLMs' performance, indicating its limitations for the subjective task of visual emotion analysis.

Yuki Hirakawa, Takashi Wada, Ryotaro Shimizu et al.

Given a person image and a garment image, image-based Virtual Try-ON (VTON) synthesizes a try-on image of the person wearing the target garment. As VTON systems become increasingly important in practical applications such as fashion e-commerce, reliable evaluation of their outputs has emerged as a critical challenge. In real-world scenarios, ground-truth images of the same person wearing the target garment are typically unavailable, making reference-based evaluation impractical. Moreover, widely used distribution-level metrics such as Fréchet Inception Distance and Kernel Inception Distance measure dataset-level similarity and fail to reflect the perceptual quality of individual generated images. To address these limitations, we propose Image Quality Assessment for Virtual Try-On (VTON-IQA), a reference-free framework for human-aligned, image-level quality assessment without requiring ground-truth images. To model human perceptual judgments, we construct VTON-QBench, a large-scale human-annotated benchmark comprising 62,688 try-on images generated by 14 representative VTON models and 431,800 quality annotations collected from 13,838 qualified annotators. To the best of our knowledge, this is the largest dataset to date for human subjective evaluation in virtual try-on. Evaluating virtual try-on quality requires verifying both garment fidelity and the preservation of person-specific details. To explicitly model such interactions, we introduce an Interleaved Cross-Attention module that extends standard transformer blocks by inserting a cross-attention layer between self-attention and MLP in the latter blocks. Extensive experiments show that VTON-IQA achieves reliable human-aligned image-level quality prediction. Moreover, we conduct a comprehensive benchmark evaluation of 14 representative VTON models using VTON-IQA.

Kosuke Sakurai, Tatsuya Ishii, Ryotaro Shimizu et al.

In recent years, considerable research has been conducted on vision-language models that handle both image and text data; these models are being applied to diverse downstream tasks, such as "image-related chat," "image recognition by instruction," and "answering visual questions." Vision-language models (VLMs), such as Contrastive Language-Image Pre-training (CLIP), are also high-performance image classifiers that are being developed into domain adaptation methods that can utilize language information to extend into unseen domains. However, because these VLMs embed images as a single point in a unified embedding space, there is room for improvement in the classification accuracy. Therefore, in this study, we proposed the Latent Augmentation using Regional Embedding (LARE), which embeds the image as a region in the unified embedding space learned by the VLM. By sampling the augmented image embeddings from within this latent region, LARE enables data augmentation to various unseen domains, not just to specific unseen domains. LARE achieves robust image classification for domains in and out using augmented image embeddings to fine-tune VLMs. We demonstrate that LARE outperforms previous fine-tuning models in terms of image classification accuracy on three benchmarks. We also demonstrate that LARE is a more robust and general model that is valid under multiple conditions, such as unseen domains, small amounts of data, and imbalanced data.

Ryotaro Shimizu, Takuma Nakamura, Masayuki Goto

In this study, we propose a technology called the Fashion Intelligence System based on the visual-semantic embedding (VSE) model to quantify abstract and complex expressions unique to fashion, such as ''casual,'' ''adult-casual,'' and ''office-casual,'' and to support users' understanding of fashion. However, the existing VSE model does not support the situations in which the image is composed of multiple parts such as hair, tops, pants, skirts, and shoes. We propose partial VSE, which enables sensitive learning for each part of the fashion coordinates. The proposed model partially learns embedded representations. This helps retain the various existing practical functionalities and enables image-retrieval tasks in which changes are made only to the specified parts and image reordering tasks that focus on the specified parts. This was not possible with conventional models. Based on both the qualitative and quantitative evaluation experiments, we show that the proposed model is superior to conventional models without increasing the computational complexity.

Kosuke Sakurai, Ryotaro Shimizu, Masayuki Goto

Segment Anything Model (SAM) represents a large-scale segmentation model that enables powerful zero-shot capabilities with flexible prompts. While SAM can segment any object in zero-shot, it requires user-provided prompts for each target image and does not attach any label information to masks. Few-shot segmentation models addressed these issues by inputting annotated reference images as prompts to SAM and can segment specific objects in target images without user-provided prompts. Previous SAM-based few-shot segmentation models only use annotated reference images as prompts, resulting in limited accuracy due to a lack of reference information. In this paper, we propose a novel few-shot segmentation model, Vision and Language reference Prompt into SAM (VLP-SAM), that utilizes the visual information of the reference images and the semantic information of the text labels by inputting not only images but also language as reference information. In particular, VLP-SAM is a simple and scalable structure with minimal learnable parameters, which inputs prompt embeddings with vision-language information into SAM using a multimodal vision-language model. To demonstrate the effectiveness of VLP-SAM, we conducted experiments on the PASCAL-5i and COCO-20i datasets, and achieved high performance in the few-shot segmentation task, outperforming the previous state-of-the-art model by a large margin (6.3% and 9.5% in mIoU, respectively). Furthermore, VLP-SAM demonstrates its generality in unseen objects that are not included in the training data. Our code is available at https://github.com/kosukesakurai1/VLP-SAM.

Shijian Wang, Linxin Song, Jieyu Zhang et al.

Current multimodal language model (MLM) training approaches overlook the influence of instruction templates. Previous research deals with this problem by leveraging hand-crafted or model-generated templates, failing to investigate the scaling effect of instruction templates on MLM training. In this work, we propose a programmatic instruction template generator capable of producing over 15K unique instruction templates by filling randomly sampled positional synonyms into weighted sampled meta templates, enabling us to comprehensively explore MLM's performance across various template scales in the training process. Our investigation into scaling instruction templates for MLM training demonstrates that MLM capabilities do not consistently improve with increasing template scale. Instead, optimal performance is achieved at a medium template scale. Models trained with data augmented at the optimal template scale achieve performance gains of up to 10% over those trained on the original data and achieve the best overall performance compared with the similar-scale MLMs tuned on at most 75 times the scale of our augmented dataset. The code will be publicly available at https://github.com/shijian2001/TemplateScaling.

Masanari Kimura, Ryotaro Shimizu, Yuki Hirakawa et al.

Conventional machine learning algorithms have traditionally been designed under the assumption that input data follows a vector-based format, with an emphasis on vector-centric paradigms. However, as the demand for tasks involving set-based inputs has grown, there has been a paradigm shift in the research community towards addressing these challenges. In recent years, the emergence of neural network architectures such as Deep Sets and Transformers has presented a significant advancement in the treatment of set-based data. These architectures are specifically engineered to naturally accommodate sets as input, enabling more effective representation and processing of set structures. Consequently, there has been a surge of research endeavors dedicated to exploring and harnessing the capabilities of these architectures for various tasks involving the approximation of set functions. This comprehensive survey aims to provide an overview of the diverse problem settings and ongoing research efforts pertaining to neural networks that approximate set functions. By delving into the intricacies of these approaches and elucidating the associated challenges, the survey aims to equip readers with a comprehensive understanding of the field. Through this comprehensive perspective, we hope that researchers can gain valuable insights into the potential applications, inherent limitations, and future directions of set-based neural networks. Indeed, from this survey we gain two insights: i) Deep Sets and its variants can be generalized by differences in the aggregation function, and ii) the behavior of Deep Sets is sensitive to the choice of the aggregation function. From these observations, we show that Deep Sets, one of the well-known permutation-invariant neural networks, can be generalized in the sense of a quasi-arithmetic mean.

Linxin Song, Xuwei Ding, Jieyu Zhang et al.

Large language models (LLMs) possess impressive linguistic capabilities but often fail to faithfully retain factual knowledge, leading to hallucinations and unreliable outputs. Understanding LLMs' knowledge deficiencies by exhaustively evaluating against full-scale knowledge bases is computationally prohibitive, especially for closed-weight models. We propose stochastic error ascent (SEA), a scalable and efficient framework for discovering knowledge deficiencies (errors) in closed-weight LLMs under a strict query budget. Rather than naively probing all knowledge candidates, SEA formulates error discovery as a stochastic optimization process: it iteratively retrieves new high-error candidates by leveraging the semantic similarity to previously observed failures. To further enhance search efficiency and coverage, SEA employs hierarchical retrieval across document and paragraph levels, and constructs a relation directed acyclic graph to model error propagation and identify systematic failure modes. Empirically, SEA uncovers 40.7x more knowledge errors than Automated Capability Discovery and 26.7% more than AutoBencher, while reducing the cost-per-error by 599x and 9x, respectively. Human evaluation confirms the high quality of generated questions, while ablation and convergence analyses validate the contribution of each component in SEA. Further analysis on the discovered errors reveals correlated failure patterns across LLM families and recurring deficits, highlighting the need for better data coverage and targeted fine-tuning in future LLM development.

Hiroki Naganuma, Kotaro Yoshida, Laura Gomezjurado Gonzalez et al.

Model editing techniques, particularly task arithmetic using task vectors, have shown promise in efficiently modifying pre-trained models through arithmetic operations like task addition and negation. Despite computational advantages, these methods may inadvertently affect model fairness, creating risks in sensitive applications like hate speech detection. However, the fairness implications of task arithmetic remain largely unexplored, presenting a critical gap in the existing literature. We systematically examine how manipulating task vectors affects fairness metrics, including Demographic Parity and Equalized Odds. To rigorously assess these effects, we benchmark task arithmetic against full fine-tuning, a costly but widely used baseline, and Low-Rank Adaptation (LoRA), a prevalent parameter-efficient fine-tuning method. Additionally, we explore merging task vectors from models fine-tuned on demographic subgroups vulnerable to hate speech, investigating whether fairness outcomes can be controlled by adjusting task vector coefficients, potentially enabling tailored model behavior. Our results offer novel insights into the fairness implications of model editing and establish a foundation for fairness-aware and responsible model editing practices.

Yuya Yoshikawa, Ryotaro Shimizu, Takahiro Kawashima et al.

In scenarios requiring both prediction and explanation efficiency for image classification, self-explaining models that perform both tasks in a single inference are effective. However, their training incurs substantial labeling and computational costs. This study aims to tackle the issue by proposing a method to transfer the visual explainability of self-explaining models, learned in a source domain, to a target domain based on a task arithmetic framework. Specifically, we construct a self-explaining model by extending image classifiers based on a vision-language pretrained model. We then define an \emph{explainability vector} as the difference between model parameters trained on the source domain with and without explanation supervision. Based on the task arithmetic framework, we impart explainability to a model trained only on the prediction task in the target domain by applying the explainability vector. Experimental results on various image classification datasets demonstrate that, except for transfers between some less-related domains, visual explainability can be successfully transferred from source to target domains, improving explanation quality in the target domain without sacrificing classification accuracy. Furthermore, we show that the explainability vector learned on a large and diverse dataset like ImageNet, extended with explanation supervision, exhibits universality and robustness, improving explanation quality on nine out of ten different target datasets. We also find that the explanation quality achieved with a single model inference is comparable to that of Kernel SHAP, which requires 150 model inferences.

Shijian Wang, Linxin Song, Ryotaro Shimizu et al.

Zero-shot domain-specific image classification is challenging in classifying real images without ground-truth in-domain training examples. Recent research involved knowledge from texts with a text-to-image model to generate in-domain training images in zero-shot scenarios. However, existing methods heavily rely on simple prompt strategies, limiting the diversity of synthetic training images, thus leading to inferior performance compared to real images. In this paper, we propose AttrSyn, which leverages large language models to generate attributed prompts. These prompts allow for the generation of more diverse attributed synthetic images. Experiments for zero-shot domain-specific image classification on two fine-grained datasets show that training with synthetic images generated by AttrSyn significantly outperforms CLIP's zero-shot classification under most situations and consistently surpasses simple prompt strategies.

Kotaro Yoshida, Yuji Naraki, Takafumi Horie et al.

Model merging has emerged as an efficient and flexible paradigm for multi-task learning, with numerous methods being proposed in recent years. However, these state-of-the-art techniques are typically evaluated on benchmark suites that are highly favorable to model merging, and their robustness in more realistic settings remains largely unexplored. In this work, we first investigate the vulnerabilities of model-merging methods and pinpoint the source-model characteristics that critically underlie them. Specifically, we identify two factors that are particularly harmful to the merging process: (1) disparities in task vector norms, and (2) the low confidence of the source models. To address this issue, we propose DisTaC (Distillation for Task vector Conditioning), a novel method that pre-conditions these problematic task vectors before the merge. DisTaC leverages knowledge distillation to adjust a task vector's norm and increase source-model confidence while preserving its essential task-specific knowledge. Our extensive experiments demonstrate that by pre-conditioning task vectors with DisTaC, state-of-the-art merging techniques can successfully integrate models exhibiting the harmful traits -- where they would otherwise fail -- achieving significant performance gains.

Takashi Wada, Yuki Hirakawa, Ryotaro Shimizu et al.

We propose new static word embeddings optimised for sentence semantic representation. We first extract word embeddings from a pre-trained Sentence Transformer, and improve them with sentence-level principal component analysis, followed by either knowledge distillation or contrastive learning. During inference, we represent sentences by simply averaging word embeddings, which requires little computational cost. We evaluate models on both monolingual and cross-lingual tasks and show that our model substantially outperforms existing static models on sentence semantic tasks, and even surpasses a basic Sentence Transformer model (SimCSE) on a text embedding benchmark. Lastly, we perform a variety of analyses and show that our method successfully removes word embedding components that are not highly relevant to sentence semantics, and adjusts the vector norms based on the influence of words on sentence semantics.

Yuki Hirakawa, Ryotaro Shimizu

Constructing dataset for fashion style recognition is challenging due to the inherent subjectivity and ambiguity of style concepts. Recent advances in text-to-image models have facilitated generative data augmentation by synthesizing images from labeled data, yet existing methods based solely on class names or reference captions often fail to balance visual diversity and style consistency. In this work, we propose \textbf{Masked Language Prompting (MLP)}, a novel prompting strategy that masks selected words in a reference caption and leverages large language models to generate diverse yet semantically coherent completions. This approach preserves the structural semantics of the original caption while introducing attribute-level variations aligned with the intended style, enabling style-consistent and diverse image generation without fine-tuning. Experimental results on the FashionStyle14 dataset demonstrate that our MLP-based augmentation consistently outperforms class-name and caption-based baselines, validating its effectiveness for fashion style recognition under limited supervision.

Qice Qin, Yuki Hirakawa, Ryotaro Shimizu et al.

Image generation in the fashion domain has predominantly focused on preserving body characteristics or following input prompts, but little attention has been paid to improving the inherent fashionability of the output images. This paper presents a novel diffusion model-based approach that generates fashion images with improved fashionability while maintaining control over key attributes. Key components of our method include: 1) fashionability enhancement, which ensures that the generated images are more fashionable than the input; 2) preservation of body characteristics, encouraging the generated images to maintain the original shape and proportions of the input; and 3) automatic fashion optimization, which does not rely on manual input or external prompts. We also employ two methods to collect training data for guidance while generating and evaluating the images. In particular, we rate outfit images using fashionability scores annotated by multiple fashion experts through OpenSkill-based and five critical aspect-based pairwise comparisons. These methods provide complementary perspectives for assessing and improving the fashionability of the generated images. The experimental results show that our approach outperforms the baseline Fashion++ in generating images with superior fashionability, demonstrating its effectiveness in producing more stylish and appealing fashion images.

Yuya Yoshikawa, Masanari Kimura, Ryotaro Shimizu et al.

Techniques that explain the predictions of black-box machine learning models are crucial to make the models transparent, thereby increasing trust in AI systems. The input features to the models often have a nested structure that consists of high- and low-level features, and each high-level feature is decomposed into multiple low-level features. For such inputs, both high-level feature attributions (HiFAs) and low-level feature attributions (LoFAs) are important for better understanding the model's decision. In this paper, we propose a model-agnostic local explanation method that effectively exploits the nested structure of the input to estimate the two-level feature attributions simultaneously. A key idea of the proposed method is to introduce the consistency property that should exist between the HiFAs and LoFAs, thereby bridging the separate optimization problems for estimating them. Thanks to this consistency property, the proposed method can produce HiFAs and LoFAs that are both faithful to the black-box models and consistent with each other, using a smaller number of queries to the models. In experiments on image classification in multiple instance learning and text classification using language models, we demonstrate that the HiFAs and LoFAs estimated by the proposed method are accurate, faithful to the behaviors of the black-box models, and provide consistent explanations.

Yuji Naraki, Ryosuke Yamaki, Yoshikazu Ikeda et al.

In the field of Natural Language Processing (NLP), Named Entity Recognition (NER) is recognized as a critical technology, employed across a wide array of applications. Traditional methodologies for annotating datasets for NER models are challenged by high costs and variations in dataset quality. This research introduces a novel hybrid annotation approach that synergizes human effort with the capabilities of Large Language Models (LLMs). This approach not only aims to ameliorate the noise inherent in manual annotations, such as omissions, thereby enhancing the performance of NER models, but also achieves this in a cost-effective manner. Additionally, by employing a label mixing strategy, it addresses the issue of class imbalance encountered in LLM-based annotations. Through an analysis across multiple datasets, this method has been consistently shown to provide superior performance compared to traditional annotation methods, even under constrained budget conditions. This study illuminates the potential of leveraging LLMs to improve dataset quality, introduces a novel technique to mitigate class imbalances, and demonstrates the feasibility of achieving high-performance NER in a cost-effective way.