This study assesses the ability of state-of-the-art large language models (LLMs) including GPT-3.5, GPT-4, Falcon, and LLaMA 2 to identify patients with mild cognitive impairment (MCI) from discharge summaries and examines instances where the models' responses were misaligned with their reasoning. Utilizing the MIMIC-IV v2.2 database, we focused on a cohort aged 65 and older, verifying MCI diagnoses against ICD codes and expert evaluations. The data was partitioned into training, validation, and testing sets in a 7:2:1 ratio for model fine-tuning and evaluation, with an additional metastatic cancer dataset from MIMIC III used to further assess reasoning consistency. GPT-4 demonstrated superior interpretative capabilities, particularly in response to complex prompts, yet displayed notable response-reasoning inconsistencies. In contrast, open-source models like Falcon and LLaMA 2 achieved high accuracy but lacked explanatory reasoning, underscoring the necessity for further research to optimize both performance and interpretability. The study emphasizes the significance of prompt engineering and the need for further exploration into the unexpected reasoning-response misalignment observed in GPT-4. The results underscore the promise of incorporating LLMs into healthcare diagnostics, contingent upon methodological advancements to ensure accuracy and clinical coherence of AI-generated outputs, thereby improving the trustworthiness of LLMs for medical decision-making.
Pre-trained Vision Language Models (VLMs) have demonstrated notable progress in various zero-shot tasks, such as classification and retrieval. Despite their performance, because improving performance on new tasks requires task-specific knowledge, their adaptation is essential. While labels are needed for the adaptation, acquiring them is typically expensive. To overcome this challenge, active learning, a method of achieving a high performance by obtaining labels for a small number of samples from experts, has been studied. Active learning primarily focuses on selecting unlabeled samples for labeling and leveraging them to train models. In this study, we pose the question, "how can the pre-trained VLMs be adapted under the active learning framework?" In response to this inquiry, we observe that (1) simply applying a conventional active learning framework to pre-trained VLMs even may degrade performance compared to random selection because of the class imbalance in labeling candidates, and (2) the knowledge of VLMs can provide hints for achieving the balance before labeling. Based on these observations, we devise a novel active learning framework for VLMs, denoted as PCB. To assess the effectiveness of our approach, we conduct experiments on seven different real-world datasets, and the results demonstrate that PCB surpasses conventional active learning and random sampling methods.
The presence of noisy labels in a training dataset can significantly impact the performance of machine learning models. To tackle this issue, researchers have explored methods for Learning with Noisy Labels to identify clean samples and reduce the influence of noisy labels. However, constraining the influence of a certain portion of the training dataset can result in a reduction in overall generalization performance. To alleviate this, recent studies have considered the careful utilization of noisy labels by leveraging huge computational resources. Therefore, the increasing training cost necessitates a reevaluation of efficiency. In other areas of research, there has been a focus on developing fine-tuning techniques for large pre-trained models that aim to achieve both high generalization performance and efficiency. However, these methods have mainly concentrated on clean datasets, and there has been limited exploration of the noisy label scenario. In this research, our aim is to find an appropriate way to fine-tune pre-trained models for noisy labeled datasets. To achieve this goal, we investigate the characteristics of pre-trained models when they encounter noisy datasets. Through empirical analysis, we introduce a novel algorithm called TURN, which robustly and efficiently transfers the prior knowledge of pre-trained models. The algorithm consists of two main steps: (1) independently tuning the linear classifier to protect the feature extractor from being distorted by noisy labels, and (2) reducing the noisy label ratio and fine-tuning the entire model based on the noise-reduced dataset to adapt it to the target dataset. The proposed algorithm has been extensively tested and demonstrates efficient yet improved denoising performance on various benchmarks compared to previous methods.
Structured pruning methods have proven effective in reducing the model size and accelerating inference speed in various network architectures such as Transformers. Despite the versatility of encoder-decoder models in numerous NLP tasks, the structured pruning methods on such models are relatively less explored compared to encoder-only models. In this study, we investigate the behavior of the structured pruning of the encoder-decoder models in the decoupled pruning perspective of the encoder and decoder component, respectively. Our findings highlight two insights: (1) the number of decoder layers is the dominant factor of inference speed, and (2) low sparsity in the pruned encoder network enhances generation quality. Motivated by these findings, we propose a simple and effective framework, NASH, that narrows the encoder and shortens the decoder networks of encoder-decoder models. Extensive experiments on diverse generation and inference tasks validate the effectiveness of our method in both speedup and output quality.
Class imbalance problems frequently occur in real-world tasks, and conventional deep learning algorithms are well known for performance degradation on imbalanced training datasets. To mitigate this problem, many approaches have aimed to balance among given classes by re-weighting or re-sampling training samples. These re-balancing methods increase the impact of minority classes and reduce the influence of majority classes on the output of models. However, the extracted representations may be of poor quality owing to the limited number of minority samples. To handle this restriction, several methods have been developed that increase the representations of minority samples by leveraging the features of the majority samples. Despite extensive recent studies, no deep analysis has been conducted on determination of classes to be augmented and strength of augmentation has been conducted. In this study, we first investigate the correlation between the degree of augmentation and class-wise performance, and find that the proper degree of augmentation must be allocated for each class to mitigate class imbalance problems. Motivated by this finding, we propose a simple and efficient novel curriculum, which is designed to find the appropriate per-class strength of data augmentation, called CUDA: CUrriculum of Data Augmentation for long-tailed recognition. CUDA can simply be integrated into existing long-tailed recognition methods. We present the results of experiments showing that CUDA effectively achieves better generalization performance compared to the state-of-the-art method on various imbalanced datasets such as CIFAR-100-LT, ImageNet-LT, and iNaturalist 2018.
Improperly constructed datasets can result in inaccurate inferences. For instance, models trained on biased datasets perform poorly in terms of generalization (i.e., dataset bias). Recent debiasing techniques have successfully achieved generalization performance by underestimating easy-to-learn samples (i.e., bias-aligned samples) and highlighting difficult-to-learn samples (i.e., bias-conflicting samples). However, these techniques may fail owing to noisy labels, because the trained model recognizes noisy labels as difficult-to-learn and thus highlights them. In this study, we find that earlier approaches that used the provided labels to quantify difficulty could be affected by the small proportion of noisy labels. Furthermore, we find that running denoising algorithms before debiasing is ineffective because denoising algorithms reduce the impact of difficult-to-learn samples, including valuable bias-conflicting samples. Therefore, we propose an approach called denoising after entropy-based debiasing, i.e., DENEB, which has three main stages. (1) The prejudice model is trained by emphasizing (bias-aligned, clean) samples, which are selected using a Gaussian Mixture Model. (2) Using the per-sample entropy from the output of the prejudice model, the sampling probability of each sample that is proportional to the entropy is computed. (3) The final model is trained using existing denoising algorithms with the mini-batches constructed by following the computed sampling probability. Compared to existing debiasing and denoising algorithms, our method achieves better debiasing performance on multiple benchmarks.
The performance of deep neural networks is strongly influenced by the training dataset setup. In particular, when attributes having a strong correlation with the target attribute are present, the trained model can provide unintended prejudgments and show significant inference errors (i.e., the dataset bias problem). Various methods have been proposed to mitigate dataset bias, and their emphasis is on weakly correlated samples, called bias-conflicting samples. These methods are based on explicit bias labels involving human or empirical correlation metrics (e.g., training loss). However, such metrics require human costs or have insufficient theoretical explanation. In this study, we propose a debiasing algorithm, called PGD (Per-sample Gradient-based Debiasing), that comprises three steps: (1) training a model on uniform batch sampling, (2) setting the importance of each sample in proportion to the norm of the sample gradient, and (3) training the model using importance-batch sampling, whose probability is obtained in step (2). Compared with existing baselines for various synthetic and real-world datasets, the proposed method showed state-of-the-art accuracy for a the classification task. Furthermore, we describe theoretical understandings about how PGD can mitigate dataset bias.
In this paper, we study the problem of unsupervised domain adaptation that aims at obtaining a prediction model for the target domain using labeled data from the source domain and unlabeled data from the target domain. There exists an array of recent research based on the idea of extracting features that are not only invariant for both domains but also provide high discriminative power for the target domain. In this paper, we propose an idea of empowering the discriminativeness: Adding a new, artificial class and training the model on the data together with the GAN-generated samples of the new class. The trained model based on the new class samples is capable of extracting the features that are more discriminative by repositioning data of current classes in the target domain and therefore drawing the decision boundaries more effectively. Our idea is highly generic so that it is compatible with many existing methods such as DANN, VADA, and DIRT-T. We conduct various experiments for the standard data commonly used for the evaluation of unsupervised domain adaptations and demonstrate that our algorithm achieves the SOTA performance for many scenarios.