Zero-shot text learning enables text classifiers to handle unseen classes efficiently, alleviating the need for task-specific training data. A simple approach often relies on comparing embeddings of query (text) to those of potential classes. However, the embeddings of a simple query sometimes lack rich contextual information, which hinders the classification performance. Traditionally, this has been addressed by improving the embedding model with expensive training. We introduce QZero, a novel training-free knowledge augmentation approach that reformulates queries by retrieving supporting categories from Wikipedia to improve zero-shot text classification performance. Our experiments across six diverse datasets demonstrate that QZero enhances performance for state-of-the-art static and contextual embedding models without the need for retraining. Notably, in News and medical topic classification tasks, QZero improves the performance of even the largest OpenAI embedding model by at least 5% and 3%, respectively. Acting as a knowledge amplifier, QZero enables small word embedding models to achieve performance levels comparable to those of larger contextual models, offering the potential for significant computational savings. Additionally, QZero offers meaningful insights that illuminate query context and verify topic relevance, aiding in understanding model predictions. Overall, QZero improves embedding-based zero-shot classifiers while maintaining their simplicity. This makes it particularly valuable for resource-constrained environments and domains with constantly evolving information.
In this paper, we develop a decision support system for the hierarchical text classification. We consider text collections with a fixed hierarchical structure of topics given by experts in the form of a tree. The system sorts the topics by relevance to a given document. The experts choose one of the most relevant topics to finish the classification. We propose a weighted hierarchical similarity function to calculate topic relevance. The function calculates the similarity of a document and a tree branch. The weights in this function determine word importance. We use the entropy of words to estimate the weights. The proposed hierarchical similarity function formulates a joint hierarchical thematic classification probability model of the document topics, parameters, and hyperparameters. The variational Bayesian inference gives a closed-form EM algorithm. The EM algorithm estimates the parameters and calculates the probability of a topic for a given document. Compared to hierarchical multiclass SVM, hierarchical PLSA with adaptive regularization, and hierarchical naive Bayes, the weighted hierarchical similarity function has better improvement in ranking accuracy in an abstract collection of a major conference EURO and a website collection of industrial companies.
Recent evaluations of cross-domain text classification models aim to measure the ability of a model to obtain domain-invariant performance in a target domain given labeled samples in a source domain. The primary strategy for this evaluation relies on assumed differences between source domain samples and target domain samples in benchmark datasets. This evaluation strategy fails to account for the similarity between source and target domains, and may mask when models fail to transfer learning to specific target samples which are highly dissimilar from the source domain. We introduce Depth $F_1$, a novel cross-domain text classification performance metric. Designed to be complementary to existing classification metrics such as $F_1$, Depth $F_1$ measures how well a model performs on target samples which are dissimilar from the source domain. We motivate this metric using standard cross-domain text classification datasets and benchmark several recent cross-domain text classification models, with the goal of enabling in-depth evaluation of the semantic generalizability of cross-domain text classification models.
Despite their success, Large-Language Models (LLMs) still face criticism as their lack of interpretability limits their controllability and reliability. Traditional post-hoc interpretation methods, based on attention and gradient-based analysis, offer limited insight into the model's decision-making processes. In the image field, Concept-based models have emerged as explainable-by-design architectures, employing human-interpretable features as intermediate representations. However, these methods have not been yet adapted to textual data, mainly because they require expensive concept annotations, which are impractical for real-world text data. This paper addresses this challenge by proposing a self-supervised Interpretable Concept Embedding Models (ICEMs). We leverage the generalization abilities of LLMs to predict the concepts labels in a self-supervised way, while we deliver the final predictions with an interpretable function. The results of our experiments show that ICEMs can be trained in a self-supervised way achieving similar performance to fully supervised concept-based models and end-to-end black-box ones. Additionally, we show that our models are (i) interpretable, offering meaningful logical explanations for their predictions; (ii) interactable, allowing humans to modify intermediate predictions through concept interventions; and (iii) controllable, guiding the LLMs' decoding process to follow a required decision-making path.
While fine-tuning of pre-trained language models generally helps to overcome the lack of labelled training samples, it also displays model performance instability. This instability mainly originates from randomness in initialisation or data shuffling. To address this, researchers either modify the training process or augment the available samples, which typically results in increased computational costs. We propose a new mitigation strategy, called Delayed Ensemble with Noisy Interpolation (DENI), that leverages the strengths of ensembling, noise regularisation and model interpolation, while retaining computational efficiency. We compare DENI with 9 representative mitigation strategies across 3 models, 4 tuning strategies and 7 text classification datasets. We show that: 1) DENI outperforms the best performing mitigation strategy (Ensemble), while using only a fraction of its cost; 2) the mitigation strategies are beneficial for parameter-efficient fine-tuning (PEFT) methods, outperforming full fine-tuning in specific cases; and 3) combining DENI with data augmentation often leads to even more effective instability mitigation.
Large Language Models (LLMs) changed the way we design and interact with software systems. Their ability to process and extract information from text has drastically improved productivity in a number of routine tasks. Developers that want to include these models in their software stack, however, face a dreadful challenge: debugging LLMs' inconsistent behavior across minor variations of the prompt. We therefore introduce two metrics for classification tasks, namely sensitivity and consistency, which are complementary to task performance. First, sensitivity measures changes of predictions across rephrasings of the prompt, and does not require access to ground truth labels. Instead, consistency measures how predictions vary across rephrasings for elements of the same class. We perform an empirical comparison of these metrics on text classification tasks, using them as guideline for understanding failure modes of the LLM. Our hope is that sensitivity and consistency will be helpful to guide prompt engineering and obtain LLMs that balance robustness with performance.
Mixture-of-Experts (MoE) has been demonstrated as an efficient method to scale up models. By dynamically and sparsely selecting activated experts, MoE can effectively reduce computational costs. Despite the success, we observe that many tokens in the MoE models have uncertain routing results. These tokens have nearly equal scores for choosing each expert, and we demonstrate that this uncertainty can lead to incorrect selections. Inspired by the Global Workspace Theory (GWT), we propose a new fine-tuning method, GW-MoE, to address this issue. The core idea is to broadcast the uncertain tokens across experts during fine-tuning. Therefore, these tokens can acquire the necessary knowledge from any expert during inference and become less sensitive to the choice. GW-MoE does not introduce additional inference overhead. We validate that GW can mitigate the uncertain problem and consistently improve in different tasks (text classification, question answering, summarization, code generation, and mathematical problem solving) and model sizes (650M and 8B parameters).
In the context of text classification, the financial burden of annotation exercises for creating training data is a critical issue. Active learning techniques, particularly those rooted in uncertainty sampling, offer a cost-effective solution by pinpointing the most instructive samples for manual annotation. Similarly, Large Language Models (LLMs) such as GPT-3.5 provide an alternative for automated annotation but come with concerns regarding their reliability. This study introduces a novel methodology that integrates human annotators and LLMs within an Active Learning framework. We conducted evaluations on three public datasets. IMDB for sentiment analysis, a Fake News dataset for authenticity discernment, and a Movie Genres dataset for multi-label classification.The proposed framework integrates human annotation with the output of LLMs, depending on the model uncertainty levels. This strategy achieves an optimal balance between cost efficiency and classification performance. The empirical results show a substantial decrease in the costs associated with data annotation while either maintaining or improving model accuracy.
The widespread popularity of Large Language Models (LLMs), partly due to their unique ability to perform in-context learning, has also brought to light the importance of ethical and safety considerations when deploying these pre-trained models. In this work, we focus on investigating machine unlearning for LLMs motivated by data protection regulations. In contrast to the growing literature on fine-tuning methods to achieve unlearning, we focus on a comparatively lightweight alternative called soft prompting to realize the unlearning of a subset of training data. With losses designed to enforce forgetting as well as utility preservation, our framework \textbf{S}oft \textbf{P}rompting for \textbf{U}n\textbf{l}earning (SPUL) learns prompt tokens that can be appended to an arbitrary query to induce unlearning of specific examples at inference time without updating LLM parameters. We conduct a rigorous evaluation of the proposed method and our results indicate that SPUL can significantly improve the trade-off between utility and forgetting in the context of text classification and question answering with LLMs. We further validate our method using multiple LLMs to highlight the scalability of our framework and provide detailed insights into the choice of hyperparameters and the influence of the size of unlearning data. Our implementation is available at \url{https://github.com/karuna-bhaila/llm_unlearning}.
For extremely weak-supervised text classification, pioneer research generates pseudo labels by mining texts similar to the class names from the raw corpus, which may end up with very limited or even no samples for the minority classes. Recent works have started to generate the relevant texts by prompting LLMs using the class names or definitions; however, there is a high risk that LLMs cannot generate in-distribution (i.e., similar to the corpus where the text classifier will be applied) data, leading to ungeneralizable classifiers. In this paper, we combine the advantages of these two approaches and propose to bridge the gap via a novel framework, \emph{text grafting}, which aims to obtain clean and near-distribution weak supervision for minority classes. Specifically, we first use LLM-based logits to mine masked templates from the raw corpus, which have a high potential for data synthesis into the target minority class. Then, the templates are filled by state-of-the-art LLMs to synthesize near-distribution texts falling into minority classes. Text grafting shows significant improvement over direct mining or synthesis on minority classes. We also use analysis and case studies to comprehend the property of text grafting.
The increasing size of transformer-based models in NLP makes the question of compressing them important. In this work, we present a comprehensive analysis of factorization based model compression techniques. Specifically, we focus on comparing straightforward low-rank factorization against the recently introduced Monarch factorization, which exhibits impressive performance preservation on the GLUE benchmark. To mitigate stability issues associated with low-rank factorization of the matrices in pre-trained transformers, we introduce a staged factorization approach wherein layers are factorized one by one instead of being factorized simultaneously. Through this strategy we significantly enhance the stability and reliability of the compression process. Further, we introduce a simple block-wise low-rank factorization method, which has a close relationship to Monarch factorization. Our experiments lead to the surprising conclusion that straightforward low-rank factorization consistently outperforms Monarch factorization across both different compression ratios and six different text classification tasks.
Federated learning (FL) has obtained tremendous progress in providing collaborative training solutions for distributed data silos with privacy guarantees. However, few existing works explore a more realistic scenario where the clients hold multiple data modalities. In this paper, we aim to solve a novel challenge in multi-modal federated learning (MFL) -- modality missing -- the clients may lose part of the modalities in their local data sets. To tackle the problems, we propose a novel multi-modal federated learning method, Federated Multi-modal contrastiVe training with Pre-trained completion (FedMVP), which integrates the large-scale pre-trained models to enhance the federated training. In the proposed FedMVP framework, each client deploys a large-scale pre-trained model with frozen parameters for modality completion and representation knowledge transfer, enabling efficient and robust local training. On the server side, we utilize generated data to uniformly measure the representation similarity among the uploaded client models and construct a graph perspective to aggregate them according to their importance in the system. We demonstrate that the model achieves superior performance over two real-world image-text classification datasets and is robust to the performance degradation caused by missing modality.
Stopwords are commonly used words in a language that are often considered to be of little value in determining the meaning or significance of a document. These words occur frequently in most texts and don't provide much useful information for tasks like sentiment analysis and text classification. English, which is a high-resource language, takes advantage of the availability of stopwords, whereas low-resource Indian languages like Marathi are very limited, standardized, and can be used in available packages, but the number of available words in those packages is low. Our work targets the curation of stopwords in the Marathi language using the MahaCorpus, with 24.8 million sentences. We make use of the TF-IDF approach coupled with human evaluation to curate a strong stopword list of 400 words. We apply the stop word removal to the text classification task and show its efficacy. The work also presents a simple recipe for stopword curation in a low-resource language. The stopwords are integrated into the mahaNLP library and publicly available on https://github.com/l3cube-pune/MarathiNLP .
Text classification, an integral task in natural language processing, involves the automatic categorization of text into predefined classes. Creating supervised labeled datasets for low-resource languages poses a considerable challenge. Unlocking the language potential of low-resource languages requires robust datasets with supervised labels. However, such datasets are scarce, and the label space is often limited. In our pursuit to address this gap, we aim to optimize existing labels/datasets in different languages. This research proposes a novel perspective on Universal Cross-Lingual Text Classification, leveraging a unified model across languages. Our approach involves blending supervised data from different languages during training to create a universal model. The supervised data for a target classification task might come from different languages covering different labels. The primary goal is to enhance label and language coverage, aiming for a label set that represents a union of labels from various languages. We propose the usage of a strong multilingual SBERT as our base model, making our novel training strategy feasible. This strategy contributes to the adaptability and effectiveness of the model in cross-lingual language transfer scenarios, where it can categorize text in languages not encountered during training. Thus, the paper delves into the intricacies of cross-lingual text classification, with a particular focus on its application for low-resource languages, exploring methodologies and implications for the development of a robust and adaptable universal cross-lingual model.
Generative AI offers a simple, prompt-based alternative to fine-tuning smaller BERT-style LLMs for text classification tasks. This promises to eliminate the need for manually labeled training data and task-specific model training. However, it remains an open question whether tools like ChatGPT can deliver on this promise. In this paper, we show that smaller, fine-tuned LLMs (still) consistently and significantly outperform larger, zero-shot prompted models in text classification. We compare three major generative AI models (ChatGPT with GPT-3.5/GPT-4 and Claude Opus) with several fine-tuned LLMs across a diverse set of classification tasks (sentiment, approval/disapproval, emotions, party positions) and text categories (news, tweets, speeches). We find that fine-tuning with application-specific training data achieves superior performance in all cases. To make this approach more accessible to a broader audience, we provide an easy-to-use toolkit alongside this paper. Our toolkit, accompanied by non-technical step-by-step guidance, enables users to select and fine-tune BERT-like LLMs for any classification task with minimal technical and computational effort.
Active learning is an iterative labeling process that is used to obtain a small labeled subset, despite the absence of labeled data, thereby enabling to train a model for supervised tasks such as text classification. While active learning has made considerable progress in recent years due to improvements provided by pre-trained language models, there is untapped potential in the often neglected unlabeled portion of the data, although it is available in considerably larger quantities than the usually small set of labeled data. In this work, we investigate how self-training, a semi-supervised approach that uses a model to obtain pseudo-labels for unlabeled data, can be used to improve the efficiency of active learning for text classification. Building on a comprehensive reproduction of four previous self-training approaches, some of which are evaluated for the first time in the context of active learning or natural language processing, we introduce HAST, a new and effective self-training strategy, which is evaluated on four text classification benchmarks. Our results show that it outperforms the reproduced self-training approaches and reaches classification results comparable to previous experiments for three out of four datasets, using as little as 25% of the data. The code is publicly available at https://github.com/chschroeder/self-training-for-sample-efficient-active-learning .
Large Language Models (LLMs) are valuable for text classification, but their vulnerabilities must not be disregarded. They lack robustness against adversarial examples, so it is pertinent to understand the impacts of different types of perturbations, and assess if those attacks could be replicated by common users with a small amount of perturbations and a small number of queries to a deployed LLM. This work presents an analysis of the effectiveness, efficiency, and practicality of three different types of adversarial attacks against five different LLMs in a sentiment classification task. The obtained results demonstrated the very distinct impacts of the word-level and character-level attacks. The word attacks were more effective, but the character and more constrained attacks were more practical and required a reduced number of perturbations and queries. These differences need to be considered during the development of adversarial defense strategies to train more robust LLMs for intelligent text classification applications.
Phishing, whether through email, SMS, or malicious websites, poses a major threat to organizations by using social engineering to trick users into revealing sensitive information. It not only compromises company's data security but also incurs significant financial losses. In this paper, we investigate whether the remarkable performance of Large Language Models (LLMs) can be leveraged for particular task like text classification, particularly detecting malicious content and compare its results with state-of-the-art Deberta V3 (DeBERTa using ELECTRA-Style Pre-Training with Gradient-Disentangled Embedding Sharing) model. We systematically assess the potential and limitations of both approaches using comprehensive public datasets comprising diverse data sources such as email, HTML, URL, SMS, and synthetic data generation. Additionally, we demonstrate how LLMs can generate convincing phishing emails, making it harder to spot scams and evaluate the performance of both models in this context. Our study delves further into the challenges encountered by DeBERTa V3 during its training phases, fine-tuning methodology and transfer learning processes. Similarly, we examine the challenges associated with LLMs and assess their respective performance. Among our experimental approaches, the transformer-based DeBERTa method emerged as the most effective, achieving a test dataset (HuggingFace phishing dataset) recall (sensitivity) of 95.17% closely followed by GPT-4 providing a recall of 91.04%. We performed additional experiments with other datasets on the trained DeBERTa V3 model and LLMs like GPT 4 and Gemini 1.5. Based on our findings, we provide valuable insights into the effectiveness and robustness of these advanced language models, offering a detailed comparative analysis that can inform future research efforts in strengthening cybersecurity measures for detecting and mitigating phishing threats.
Text classification is a crucial task encountered frequently in practical scenarios, yet it is still under-explored in the era of large language models (LLMs). This study shows that LLMs are vulnerable to changes in the number and arrangement of options in text classification. Our extensive empirical analyses reveal that the key bottleneck arises from ambiguous decision boundaries and inherent biases towards specific tokens and positions. To mitigate these issues, we make the first attempt and propose a novel two-stage classification framework for LLMs. Our approach is grounded in the empirical observation that pairwise comparisons can effectively alleviate boundary ambiguity and inherent bias. Specifically, we begin with a self-reduction technique to efficiently narrow down numerous options, which contributes to reduced decision space and a faster comparison process. Subsequently, pairwise contrastive comparisons are employed in a chain-of-thought manner to draw out nuances and distinguish confusable options, thus refining the ambiguous decision boundary. Extensive experiments on four datasets (Banking77, HWU64, LIU54, and Clinic150) verify the effectiveness of our framework. Furthermore, benefitting from our framework, various LLMs can achieve consistent improvements. Our code and data are available in \url{https://github.com/Chuge0335/PC-CoT}.
Large models have demonstrated exceptional generalization capabilities in computer vision and natural language processing. Recent efforts have focused on enhancing these models with multimodal processing abilities. However, addressing the challenges posed by scenarios where one modality is absent remains a significant hurdle. In response to this issue, we propose a robust latent representation tuning method for large models. Specifically, our approach introduces a modality latent translation module to maximize the correlation between modalities, resulting in a robust representation. Following this, a newly designed fusion module is employed to facilitate information interaction between the modalities. Within this framework, common semantics are refined during training, and robust performance is achieved even in the absence of one modality. Importantly, our method maintains the frozen state of the image and text foundation models to preserve their capabilities acquired through large-scale pretraining. We conduct experiments on several public datasets, and the results underscore the effectiveness of our proposed method.