A comprehensive index of artificial intelligence and machine-learning research with AI-generated summaries, citation metrics, and direct links to papers and code.
Unknown
Trains random-forest classifiers on k-gram features from 800 paired reasoning traces to statistically detect unanswerable GSM8K problems and enable early stopping.
Amgad Muneer, Suliman Mohamed Fati
This paper compares seven machine learning classifiers for detecting cyberbullying on Twitter using a global dataset of 37,373 tweets, finding Logistic Regression achieves the best accuracy and F1 score.
Tianjing Wang, Qi Liu, Hang Shen, et al.
GraphACGAN integrates ACGAN with GNNs to improve network intrusion detection by leveraging graph structures and generating minority-class traffic.
Yung-Sung Chuang, Linlu Qiu, Cheng-Yu Hsieh, et al.
Proposes Lookback Lens, a simple hallucination detector using attention weight ratios, effective across tasks and models, and reduces hallucinations via classifier-guided decoding.
Fabrizio Sebastiani
This survey reviews machine learning approaches to automated text categorization, covering document representation, classifier construction, and evaluation.
Nathan Shone, Trần Nguyên Ngọc, Vu Dinh Phai, et al.
Proposes a nonsymmetric deep autoencoder (NDAE) for unsupervised feature learning and a stacked NDAE classifier for network intrusion detection, evaluated on KDD Cup '99 and NSL-KDD datasets.
Jessica Fridrich, Jan Kodovský
This paper introduces a novel strategy for building steganalysis detectors by assembling a rich model of noise residuals and using ensemble classifiers, achieving high detection accuracy against spatial-domain steganography.
Robi Polikar
This paper reviews ensemble-based systems in decision making, covering conditions for their superiority over single classifiers, algorithms like bagging and boosting, combination rules, and future applications.
Takaya Saito, Marc Rehmsmeier
This paper demonstrates that Precision-Recall (PRC) plots are more informative than ROC plots for evaluating binary classifiers on imbalanced datasets, as ROC plots can be misleading due to misinterpretation of specificity.
Sebastian Bach, Alexander Binder, Grégoire Montavon, et al.
Proposes Layer-Wise Relevance Propagation (LRP) to visualize pixel contributions for nonlinear classifiers, enabling interpretable heatmaps of classification decisions.
Unknown
Self-supervised Vision Transformers learn explicit semantic segmentation and excel as k-NN classifiers, unlike convolutional networks.
R. Vinayakumar, Mamoun Alazab, K. P. Soman, et al.
This paper proposes a deep neural network (DNN) based intrusion detection system that outperforms classical machine learning classifiers across multiple benchmark datasets and introduces a scalable hybrid framework, scale-hybrid-IDS-AlertNet, for rea