US corporations regularly spend millions of dollars reviewing electronically-stored documents in legal matters. Recently, attorneys apply text classification to efficiently cull massive volumes of data to identify responsive documents for use in these matters. While text classification is regularly used to reduce the discovery costs of legal matters, it also faces a perception challenge: amongst lawyers, this technology is sometimes looked upon as a "black box". Put simply, no extra information is provided for attorneys to understand why documents are classified as responsive. In recent years, explainable machine learning has emerged as an active research area. In an explainable machine learning system, predictions or decisions made by a machine learning model are human understandable. In legal 'document review' scenarios, a document is responsive, because one or more of its small text snippets are deemed responsive. In these scenarios, if these responsive snippets can be located, then attorneys could easily evaluate the model's document classification decisions - this is especially important in the field of responsible AI. Our prior research identified that predictive models created using annotated training text snippets improved the precision of a model when compared to a model created using all of a set of documents' text as training. While interesting, manually annotating training text snippets is not generally practical during a legal document review. However, small increases in precision can drastically decrease the cost of large document reviews. Automating the identification of training text snippets without human review could then make the application of training text snippet-based models a practical approach.
This paper introduces a novel approach for identifying the possible large language models (LLMs) involved in text generation. Instead of adding an additional classification layer to a base LM, we reframe the classification task as a next-token prediction task and directly fine-tune the base LM to perform it. We utilize the Text-to-Text Transfer Transformer (T5) model as the backbone for our experiments. We compared our approach to the more direct approach of utilizing hidden states for classification. Evaluation shows the exceptional performance of our method in the text classification task, highlighting its simplicity and efficiency. Furthermore, interpretability studies on the features extracted by our model reveal its ability to differentiate distinctive writing styles among various LLMs even in the absence of an explicit classifier. We also collected a dataset named OpenLLMText, containing approximately 340k text samples from human and LLMs, including GPT3.5, PaLM, LLaMA, and GPT2.
Language models (LMs) have achieved notable success in numerous NLP tasks, employing both fine-tuning and in-context learning (ICL) methods. While language models demonstrate exceptional performance, they face robustness challenges due to spurious correlations arising from imbalanced label distributions in training data or ICL exemplars. Previous research has primarily concentrated on word, phrase, and syntax features, neglecting the concept level, often due to the absence of concept labels and difficulty in identifying conceptual content in input texts. This paper introduces two main contributions. First, we employ ChatGPT to assign concept labels to texts, assessing concept bias in models during fine-tuning or ICL on test data. We find that LMs, when encountering spurious correlations between a concept and a label in training or prompts, resort to shortcuts for predictions. Second, we introduce a data rebalancing technique that incorporates ChatGPT-generated counterfactual data, thereby balancing label distribution and mitigating spurious correlations. Our method's efficacy, surpassing traditional token removal approaches, is validated through extensive testing.
In the realm of text manipulation and linguistic transformation, the question of authorship has been a subject of fascination and philosophical inquiry. Much like the Ship of Theseus paradox, which ponders whether a ship remains the same when each of its original planks is replaced, our research delves into an intriguing question: Does a text retain its original authorship when it undergoes numerous paraphrasing iterations? Specifically, since Large Language Models (LLMs) have demonstrated remarkable proficiency in both the generation of original content and the modification of human-authored texts, a pivotal question emerges concerning the determination of authorship in instances where LLMs or similar paraphrasing tools are employed to rephrase the text--i.e., whether authorship should be attributed to the original human author or the AI-powered tool. Therefore, we embark on a philosophical voyage through the seas of language and authorship to unravel this intricate puzzle. Using a computational approach, we discover that the diminishing performance in text classification models, with each successive paraphrasing iteration, is closely associated with the extent of deviation from the original author's style, thus provoking a reconsideration of the current notion of authorship.
Most pretrained language models rely on subword tokenization, which processes text as a sequence of subword tokens. However, different granularities of text, such as characters, subwords, and words, can contain different kinds of information. Previous studies have shown that incorporating multiple input granularities improves model generalization, yet very few of them outputs useful representations for each granularity. In this paper, we introduce the entanglement model, aiming to combine character and subword language models. Inspired by vision-language models, our model treats characters and subwords as separate modalities, and it generates mutually informed representations for both granularities as output. We evaluate our model on text classification, named entity recognition, POS-tagging, and character-level sequence labeling (intraword code-switching). Notably, the entanglement model outperforms its backbone language models, particularly in the presence of noisy texts and low-resource languages. Furthermore, the entanglement model even outperforms larger pre-trained models on all English sequence labeling tasks and classification tasks. We make our code publically available.
Model fusion research aims to aggregate the knowledge of multiple individual models to enhance performance by combining their weights. In this work, we study the inverse problem: investigating whether model fusion can be used to reduce unwanted knowledge. We investigate the effects of model fusion in three scenarios: the learning of shortcuts, social biases, and memorization of training data in fine-tuned language models. Through experiments covering classification and generation tasks, our analysis highlights that shared knowledge among models is enhanced during model fusion, while unshared knowledge is usually forgotten. Based on this observation, we demonstrate the potential of model fusion as a debiasing tool and showcase its efficacy in addressing privacy concerns associated with language models.
Downstream applications often require text classification models to be accurate and robust. While the accuracy of the state-of-the-art Language Models (LMs) approximates human performance, they often exhibit a drop in performance on noisy data found in the real world. This lack of robustness can be concerning, as even small perturbations in the text, irrelevant to the target task, can cause classifiers to incorrectly change their predictions. A potential solution can be the family of Prototype-Based Networks (PBNs) that classifies examples based on their similarity to prototypical examples of a class (prototypes) and has been shown to be robust to noise for computer vision tasks. In this paper, we study whether the robustness properties of PBNs transfer to text classification tasks under both targeted and static adversarial attack settings. Our results show that PBNs, as a mere architectural variation of vanilla LMs, offer more robustness compared to vanilla LMs under both targeted and static settings. We showcase how PBNs' interpretability can help us to understand PBNs' robustness properties. Finally, our ablation studies reveal the sensitivity of PBNs' robustness to how strictly clustering is done in the training phase, as tighter clustering results in less robust PBNs.
Language model (LM) prompting--a popular paradigm for solving NLP tasks--has been shown to be susceptible to miscalibration and brittleness to slight prompt variations, caused by its discriminative prompting approach, i.e., predicting the label given the input. To address these issues, we propose Gen-Z--a generative prompting framework for zero-shot text classification. GEN-Z is generative, as it measures the LM likelihood of input text, conditioned on natural language descriptions of labels. The framework is multivariate, as label descriptions allow us to seamlessly integrate additional contextual information about the labels to improve task performance. On various standard classification benchmarks, with six open-source LM families, we show that zero-shot classification with simple contextualization of the data source of the evaluation set consistently outperforms both zero-shot and few-shot baselines while improving robustness to prompt variations. Further, our approach enables personalizing classification in a zero-shot manner by incorporating author, subject, or reader information in the label descriptions.
Information Extraction (IE) stands as a cornerstone in natural language processing, traditionally segmented into distinct sub-tasks. The advent of Large Language Models (LLMs) heralds a paradigm shift, suggesting the feasibility of a singular model addressing multiple IE subtasks. In this vein, we introduce the General Information Extraction Large Language Model (GIELLM), which integrates text Classification, Sentiment Analysis, Named Entity Recognition, Relation Extraction, and Event Extraction using a uniform input-output schema. This innovation marks the first instance of a model simultaneously handling such a diverse array of IE subtasks. Notably, the GIELLM leverages the Mutual Reinforcement Effect (MRE), enhancing performance in integrated tasks compared to their isolated counterparts. Our experiments demonstrate State-of-the-Art (SOTA) results in five out of six Japanese mixed datasets, significantly surpassing GPT-3.5-Turbo. Further, an independent evaluation using the novel Text Classification Relation and Event Extraction(TCREE) dataset corroborates the synergistic advantages of MRE in text and word classification. This breakthrough paves the way for most IE subtasks to be subsumed under a singular LLM framework. Specialized fine-tune task-specific models are no longer needed.
Standard Full-Data classifiers in NLP demand thousands of labeled examples, which is impractical in data-limited domains. Few-shot methods offer an alternative, utilizing contrastive learning techniques that can be effective with as little as 20 examples per class. Similarly, Large Language Models (LLMs) like GPT-4 can perform effectively with just 1-5 examples per class. However, the performance-cost trade-offs of these methods remain underexplored, a critical concern for budget-limited organizations. Our work addresses this gap by studying the aforementioned approaches over the Banking77 financial intent detection dataset, including the evaluation of cutting-edge LLMs by OpenAI, Cohere, and Anthropic in a comprehensive set of few-shot scenarios. We complete the picture with two additional methods: first, a cost-effective querying method for LLMs based on retrieval-augmented generation (RAG), able to reduce operational costs multiple times compared to classic few-shot approaches, and second, a data augmentation method using GPT-4, able to improve performance in data-limited scenarios. Finally, to inspire future research, we provide a human expert's curated subset of Banking77, along with extensive error analysis.
Research on data generation and augmentation has been focused majorly on enhancing generation models, leaving a notable gap in the exploration and refinement of methods for evaluating synthetic data. There are several text similarity metrics within the context of generated data filtering which can impact the performance of specific Natural Language Understanding (NLU) tasks, specifically focusing on intent and sentiment classification. In this study, we propose RankAug, a text-ranking approach that detects and filters out the top augmented texts in terms of being most similar in meaning with lexical and syntactical diversity. Through experiments conducted on multiple datasets, we demonstrate that the judicious selection of filtering techniques can yield a substantial improvement of up to 35% in classification accuracy for under-represented classes.
Maintenance work orders are commonly used to document information about wind turbine operation and maintenance. This includes details about proactive and reactive wind turbine downtimes, such as preventative and corrective maintenance. However, the information contained in maintenance work orders is often unstructured and difficult to analyze, presenting challenges for decision-makers wishing to use it for optimizing operation and maintenance. To address this issue, this work compares three different approaches to calculate reliability by performance indicators from maintenance work orders. The first approach involves manual labeling of the maintenance work orders by domain experts, using the schema defined in an industrial guideline to assign the label accordingly. The second approach involves the development of a model that automatically labels the maintenance work orders using text classification methods. Through this method, we are able to achieve macro average and weighted average F1-Scores of 0.75 and 0.85 respectively. The third technique uses an AI-assisted tagging tool to tag and structure the raw maintenance information, together with a novel rule-based approach for extracting relevant maintenance work orders for failure rate calculation. In our experiments the AI-assisted tool leads to a 88% drop in tagging time in comparison to the other two approaches, while expert labeling and text classification are more accurate in KPI extraction. Overall, our findings make extracting maintenance information from maintenance work orders more efficient, enable the assessment of reliability key performance indicators and therefore support the optimization of wind turbine operation and maintenance.
Continual learning strives to ensure stability in solving previously seen tasks while demonstrating plasticity in a novel domain. Recent advances in continual learning are mostly confined to a supervised learning setting, especially in NLP domain. In this work, we consider a few-shot continual active learning setting where labeled data are inadequate, and unlabeled data are abundant but with a limited annotation budget. We exploit meta-learning and propose a method, called Meta-Continual Active Learning. This method sequentially queries the most informative examples from a pool of unlabeled data for annotation to enhance task-specific performance and tackle continual learning problems through meta-objective. Specifically, we employ meta-learning and experience replay to address inter-task confusion and catastrophic forgetting. We further incorporate textual augmentations to avoid memory over-fitting caused by experience replay and sample queries, thereby ensuring generalization. We conduct extensive experiments on benchmark text classification datasets from diverse domains to validate the feasibility and effectiveness of meta-continual active learning. We also analyze the impact of different active learning strategies on various meta continual learning models. The experimental results demonstrate that introducing randomness into sample selection is the best default strategy for maintaining generalization in meta-continual learning framework.
Pre-trained language models (PLMs) have seen tremendous success in text classification (TC) problems in the context of Natural Language Processing (NLP). In many real-world text classification tasks, the class definitions being learned do not remain constant but rather change with time - this is known as Concept Shift. Most techniques for handling concept shift rely on retraining the old classifiers with the newly labelled data. However, given the amount of training data required to fine-tune large DL models for the new concepts, the associated labelling costs can be prohibitively expensive and time consuming. In this work, we propose a reformulation, converting vanilla classification into an entailment-style problem that requires significantly less data to re-train the text classifier to adapt to new concepts. We demonstrate the effectiveness of our proposed method on both real world & synthetic datasets achieving absolute F1 gains upto 7% and 40% respectively in few-shot settings. Further, upon deployment, our solution also helped save 75% of labeling costs overall.
Recent Large Language Models (LLMs) have demonstrated remarkable capabilities in generating text that closely resembles human writing across wide range of styles and genres. However, such capabilities are prone to potential abuse, such as fake news generation, spam email creation, and misuse in academic assignments. Hence, it is essential to build automated approaches capable of distinguishing between artificially generated text and human-authored text. In this paper, we propose a simple yet efficient solution to this problem by ensembling predictions from multiple constituent LLMs. Compared to previous state-of-the-art approaches, which are perplexity-based or uses ensembles with a number of LLMs, our condensed ensembling approach uses only two constituent LLMs to achieve comparable performance. Experiments conducted on four benchmark datasets for generative text classification show performance improvements in the range of 0.5 to 100\% compared to previous state-of-the-art approaches. We also study the influence that the training data from individual LLMs have on model performance. We found that substituting commercially-restrictive Generative Pre-trained Transformer (GPT) data with data generated from other open language models such as Falcon, Large Language Model Meta AI (LLaMA2), and Mosaic Pretrained Transformers (MPT) is a feasible alternative when developing generative text detectors. Furthermore, to demonstrate zero-shot generalization, we experimented with an English essays dataset, and results suggest that our ensembling approach can handle new data effectively.
Many users reading online articles in various magazines may suffer considerable difficulty in distinguishing the implicit intents in texts. In this work, we focus on automatically recognizing the political intents of a given online newspaper by understanding the context of the text. To solve this task, we present a novel Korean text classification dataset that contains various articles. We also provide deep-learning-based text classification baseline models trained on the proposed dataset. Our dataset contains 12,000 news articles that may contain political intentions, from the politics section of six of the most representative newspaper organizations in South Korea. All the text samples are labeled simultaneously in two aspects (1) the level of political orientation and (2) the level of pro-government. To the best of our knowledge, our paper is the most large-scale Korean news dataset that contains long text and addresses multi-task classification problems. We also train recent state-of-the-art (SOTA) language models that are based on transformer architectures and demonstrate that the trained models show decent text classification performance. All the codes, datasets, and trained models are available at https://github.com/Kdavid2355/KoPolitic-Benchmark-Dataset.
Most Machine Learning research evaluates the best solutions in terms of performance. However, in the race for the best performing model, many important aspects are often overlooked when, on the contrary, they should be carefully considered. In fact, sometimes the gaps in performance between different approaches are neglectable, whereas factors such as production costs, energy consumption, and carbon footprint must take into consideration. Large Language Models (LLMs) are extensively adopted to address NLP problems in academia and industry. In this work, we present a detailed quantitative comparison of LLM and traditional approaches (e.g. SVM) on the LexGLUE benchmark, which takes into account both performance (standard indices) and alternative metrics such as timing, power consumption and cost, in a word: the carbon-footprint. In our analysis, we considered the prototyping phase (model selection by training-validation-test iterations) and in-production phases separately, since they follow different implementation procedures and also require different resources. The results indicate that very often, the simplest algorithms achieve performance very close to that of large LLMs but with very low power consumption and lower resource demands. The results obtained could suggest companies to include additional evaluations in the choice of Machine Learning (ML) solutions.
Text classification aims to effectively categorize documents into pre-defined categories. Traditional methods for text classification often rely on large amounts of manually annotated training data, making the process time-consuming and labor-intensive. To address this issue, recent studies have focused on weakly-supervised and extremely weakly-supervised settings, which require minimal or no human annotation, respectively. In previous methods of weakly supervised text classification, pseudo-training data is generated by assigning pseudo-labels to documents based on their alignment (e.g., keyword matching) with specific classes. However, these methods ignore the importance of incorporating the explanations of the generated pseudo-labels, or saliency of individual words, as additional guidance during the text classification training process. To address this limitation, we propose XAI-CLASS, a novel explanation-enhanced extremely weakly-supervised text classification method that incorporates word saliency prediction as an auxiliary task. XAI-CLASS begins by employing a multi-round question-answering process to generate pseudo-training data that promotes the mutual enhancement of class labels and corresponding explanation word generation. This pseudo-training data is then used to train a multi-task framework that simultaneously learns both text classification and word saliency prediction. Extensive experiments on several weakly-supervised text classification datasets show that XAI-CLASS outperforms other weakly-supervised text classification methods significantly. Moreover, experiments demonstrate that XAI-CLASS enhances both model performance and explainability.
Clinical note classification is a common clinical NLP task. However, annotated data-sets are scarse. Prompt-based learning has recently emerged as an effective method to adapt pre-trained models for text classification using only few training examples. A critical component of prompt design is the definition of the template (i.e. prompt text). The effect of template position, however, has been insufficiently investigated. This seems particularly important in the clinical setting, where task-relevant information is usually sparse in clinical notes. In this study we develop a keyword-optimized template insertion method (KOTI) and show how optimizing position can improve performance on several clinical tasks in a zero-shot and few-shot training setting.
Transformer-based models, specifically BERT, have propelled research in various NLP tasks. However, these models are limited to a maximum token limit of 512 tokens. Consequently, this makes it non-trivial to apply it in a practical setting with long input. Various complex methods have claimed to overcome this limit, but recent research questions the efficacy of these models across different classification tasks. These complex architectures evaluated on carefully curated long datasets perform at par or worse than simple baselines. In this work, we propose a relatively simple extension to vanilla BERT architecture called ChunkBERT that allows finetuning of any pretrained models to perform inference on arbitrarily long text. The proposed method is based on chunking token representations and CNN layers, making it compatible with any pre-trained BERT. We evaluate chunkBERT exclusively on a benchmark for comparing long-text classification models across a variety of tasks (including binary classification, multi-class classification, and multi-label classification). A BERT model finetuned using the ChunkBERT method performs consistently across long samples in the benchmark while utilizing only a fraction (6.25\%) of the original memory footprint. These findings suggest that efficient finetuning and inference can be achieved through simple modifications to pre-trained BERT models.