Kotaro Yoshida

Kotaro Yoshida, So Kuroki, Yuki Imajuku et al.

Large language models (LLMs) are increasingly used for writing and review support, but their usefulness depends on context-dependent criteria, such as expert preferences or organization-specific conventions, that are often tacit, undocumented, and difficult to elicit directly. We propose a problem setting for learning reusable natural-language rubrics from accumulated inline comments on artifacts such as human-written or LLM-generated drafts. Our method infers rubrics from these comments and iteratively refines them by observing comment-wise mismatches between rubric-conditioned predictions and reference comments. We evaluate the proposed method in real-world review settings and in controlled settings with reference rubrics. These results show that inline comments can be distilled into reusable rubrics that support comment prediction, rubric understanding, and automatic artifact revision.

Hiroki Naganuma, Ryuichiro Hataya, Kotaro Yoshida et al.

In the field of computer vision, fine-tuning pre-trained models has become a prevalent strategy for out-of-distribution (OOD) generalization tasks. Different from most prior work that has focused on advancing learning algorithms, we systematically examined how pre-trained model size, pre-training dataset size, and training strategies impact generalization and confidence calibration on downstream tasks. We evaluated 100 models across diverse pre-trained model sizes, five pre-training datasets, and five data augmentations through extensive experiments on four distribution shift datasets totaling over 120,000 GPU hours. Our results demonstrate the significant impact of pre-trained model selection, with optimal choices substantially improving OOD accuracy over algorithm improvement alone. Additionally, we find that larger models and bigger pre-training datasets not only enhance OOD performance but also improve calibration, helping to mitigate overconfidence, contrary to some prior studies that found modern deep networks to calibrate worse than classical shallow models. Our work underscores the overlooked importance of pre-trained model selection for out-of-distribution generalization and calibration.

Sora Nakai, Youssef Fadhloun, Kacem Mathlouthi et al.

Generalization remains a central yet unresolved challenge in deep learning, particularly the ability to predict a model's performance beyond its training distribution using quantities available prior to test-time evaluation. Building on the large-scale study of Jiang et al. (2020). and concerns by Dziugaite et al. (2020). about instability across training configurations, we benchmark the robustness of generalization measures beyond IID regime. We train small-to-medium models over 10,000 hyperparameter configurations and evaluate more than 40 measures computable from the trained model and the available training data alone. We significantly broaden the experimental scope along multiple axes: (i) extending the evaluation beyond the standard IID setting to include benchmarking for robustness across diverse distribution shifts, (ii) evaluating multiple architectures and training recipes, and (iii) newly incorporating calibration- and information-criteria-based measures to assess their alignment with both IID and OOD generalization. We find that distribution shifts can substantially alter the predictive performance of many generalization measures, while a smaller subset remains comparatively stable across settings.

Kotaro Yoshida, Hiroki Naganuma

Machine learning models traditionally assume that training and test data are independently and identically distributed. However, in real-world applications, the test distribution often differs from training. This problem, known as out-of-distribution (OOD) generalization, challenges conventional models. Invariant Risk Minimization (IRM) emerges as a solution that aims to identify invariant features across different environments to enhance OOD robustness. However, IRM's complexity, particularly its bi-level optimization, has led to the development of various approximate methods. Our study investigates these approximate IRM techniques, using the consistency and variance of calibration across environments as metrics to measure the invariance aimed for by IRM. Calibration, which measures the reliability of model prediction, serves as an indicator of whether models effectively capture environment-invariant features by showing how uniformly over-confident the model remains across varied environments. Through a comparative analysis of datasets with distributional shifts, we observe that Information Bottleneck-based IRM achieves consistent calibration across different environments. This observation suggests that information compression techniques, such as IB, are potentially effective in achieving model invariance. Furthermore, our empirical evidence indicates that models exhibiting consistent calibration across environments are also well-calibrated. This demonstrates that invariance and cross-environment calibration are empirically equivalent. Additionally, we underscore the necessity for a systematic approach to evaluating OOD generalization. This approach should move beyond traditional metrics, such as accuracy and F1 scores, which fail to account for the model's degree of over-confidence, and instead focus on the nuanced interplay between accuracy, calibration, and model invariance.

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.

Kotaro Yoshida, Atsushi Nitanda

Preconditioning is widely used in machine learning to accelerate convergence on the empirical risk, yet its role on the expected risk remains underexplored. In this work, we investigate how preconditioning affects feature learning and generalization performance. We first show that the input information available to the model is conveyed solely through the Gram matrix defined by the preconditioner's metric, thereby inducing a controllable spectral bias on feature learning. Concretely, instantiating the preconditioner as the $p$-th power of the input covariance matrix and within a single-index teacher model, we prove that in generalization, the exponent $p$ and the alignment between the teacher and the input spectrum are crucial factors. We further investigate how the interplay between these factors influences feature learning from three complementary perspectives: (i) Robustness to noise, (ii) Out-of-distribution generalization, and (iii) Forward knowledge transfer. Our results indicate that the learned feature representations closely mirror the spectral bias introduced by the preconditioner -- favoring components that are emphasized and exhibiting reduced sensitivity to those that are suppressed. Crucially, we demonstrate that generalization is significantly enhanced when this spectral bias is aligned with that of the teacher.

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.

Kotaro Yoshida, Konstantinos Slavakis

Invariant risk minimization (IRM) aims to enable out-of-distribution (OOD) generalization in deep learning by learning invariant representations. As IRM poses an inherently challenging bi-level optimization problem, most existing approaches -- including IRMv1 -- adopt penalty-based single-level approximations. However, empirical studies consistently show that these methods often fail to outperform well-tuned empirical risk minimization (ERM), highlighting the need for more robust IRM implementations. This work theoretically identifies a key limitation common to many IRM variants: their penalty terms are highly sensitive to limited environment diversity and over-parameterization, resulting in performance degradation. To address this issue, a novel extrapolation-based framework is proposed that enhances environmental diversity by augmenting the IRM penalty through synthetic distributional shifts. Extensive experiments -- ranging from synthetic setups to realistic, over-parameterized scenarios -- demonstrate that the proposed method consistently outperforms state-of-the-art IRM variants, validating its effectiveness and robustness.

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.