topic classification

Existing techniques for training language models can be misaligned with the truth: if we train models with imitation learning, they may reproduce errors that humans make; if we train them to generate text that humans rate highly, they may output errors that human evaluators can't detect. We propose circumventing this issue by directly finding latent knowledge inside the internal activations of a language model in a purely unsupervised way. Specifically, we introduce a method for accurately answering yes-no questions given only unlabeled model activations. It works by finding a direction in activation space that satisfies logical consistency properties, such as that a statement and its negation have opposite truth values. We show that despite using no supervision and no model outputs, our method can recover diverse knowledge represented in large language models: across 6 models and 10 question-answering datasets, it outperforms zero-shot accuracy by 4\% on average. We also find that it cuts prompt sensitivity in half and continues to maintain high accuracy even when models are prompted to generate incorrect answers. Our results provide an initial step toward discovering what language models know, distinct from what they say, even when we don't have access to explicit ground truth labels.

Variants of Naive Bayes (NB) and Support Vector Machines (SVM) are often used as baseline methods for text classification, but their performance varies greatly depending on the model variant, features used and task/dataset. We show that: (i) the inclusion of word bigram features gives consistent gains on sentiment analysis tasks; (ii) for short snippet sentiment tasks, NB actually does better than SVMs (while for longer documents the opposite result holds); (iii) a simple but novel SVM variant using NB log-count ratios as feature values consistently performs well across tasks and datasets. Based on these observations, we identify simple NB and SVM variants which outperform most published results on sentiment analysis datasets, sometimes providing a new state-of-the-art performance level.

Text classification is a foundational task in many NLP applications. Traditional text classifiers often rely on many human-designed features, such as dictionaries, knowledge bases and special tree kernels. In contrast to traditional methods, we introduce a recurrent convolutional neural network for text classification without human-designed features. In our model, we apply a recurrent structure to capture contextual information as far as possible when learning word representations, which may introduce considerably less noise compared to traditional window-based neural networks. We also employ a max-pooling layer that automatically judges which words play key roles in text classification to capture the key components in texts. We conduct experiments on four commonly used datasets. The experimental results show that the proposed method outperforms the state-of-the-art methods on several datasets, particularly on document-level datasets.