The semantic capabilities of large language models (LLMs) have the potential to enable rich analytics and reasoning over vast knowledge corpora. Unfortunately, existing systems either empirically optimize expensive LLM-powered operations with no performance guarantees, or serve a limited set of row-wise LLM operations, providing limited robustness, expressiveness and usability. We introduce semantic operators, the first formalism for declarative and general-purpose AI-based transformations based on natural language specifications (e.g., filtering, sorting, joining or aggregating records using natural language criteria). Each operator opens a rich space for execution plans, similar to relational operators. Our model specifies the expected behavior of each operator with a high-quality gold algorithm, and we develop an optimization framework that reduces cost, while providing accuracy guarantees with respect to a gold algorithm. Using this approach, we propose several novel optimizations to accelerate semantic filtering, joining, group-by and top-k operations by up to $1,000\times$. We implement semantic operators in the LOTUS system and demonstrate LOTUS' effectiveness on real, bulk-semantic processing applications, including fact-checking, biomedical multi-label classification, search, and topic analysis. We show that the semantic operator model is expressive, capturing state-of-the-art AI pipelines in a few operator calls, and making it easy to express new pipelines that match or exceed quality of recent LLM-based analytic systems by up to $170\%$, while offering accuracy guarantees. Overall, LOTUS programs match or exceed the accuracy of state-of-the-art AI pipelines for each task while running up to $3.6\times$ faster than the highest-quality baselines. LOTUS is publicly available at https://github.com/lotus-data/lotus.
Recently, various pre-trained language models (PLMs) have been proposed to prove their impressive performances on a wide range of few-shot tasks. However, limited by the unstructured prior knowledge in PLMs, it is difficult to maintain consistent performance on complex structured scenarios, such as hierarchical text classification (HTC), especially when the downstream data is extremely scarce. The main challenge is how to transfer the unstructured semantic space in PLMs to the downstream domain hierarchy. Unlike previous work on HTC which directly performs multi-label classification or uses graph neural network (GNN) to inject label hierarchy, in this work, we study the HTC problem under a few-shot setting to adapt knowledge in PLMs from an unstructured manner to the downstream hierarchy. Technically, we design a simple yet effective method named Hierarchical Iterative Conditional Random Field (HierICRF) to search the most domain-challenging directions and exquisitely crafts domain-hierarchy adaptation as a hierarchical iterative language modeling problem, and then it encourages the model to make hierarchical consistency self-correction during the inference, thereby achieving knowledge transfer with hierarchical consistency preservation. We perform HierICRF on various architectures, and extensive experiments on two popular HTC datasets demonstrate that prompt with HierICRF significantly boosts the few-shot HTC performance with an average Micro-F1 by 28.80% to 1.50% and Macro-F1 by 36.29% to 1.5% over the previous state-of-the-art (SOTA) baselines under few-shot settings, while remaining SOTA hierarchical consistency performance.
We explore methods to solve the multi-label classification task posed by the GeoLifeCLEF 2024 competition with the DS@GT team, which aims to predict the presence and absence of plant species at specific locations using spatial and temporal remote sensing data. Our approach uses frequency-domain coefficients via the Discrete Cosine Transform (DCT) to compress and pre-compute the raw input data for convolutional neural networks. We also investigate nearest neighborhood models via locality-sensitive hashing (LSH) for prediction and to aid in the self-supervised contrastive learning of embeddings through tile2vec. Our best competition model utilized geolocation features with a leaderboard score of 0.152 and a best post-competition score of 0.161. Source code and models are available at https://github.com/dsgt-kaggle-clef/geolifeclef-2024.
Multi-label ranking, which returns multiple top-ranked labels for each instance, has a wide range of applications for visual tasks. Due to its complicated setting, prior arts have proposed various measures to evaluate model performances. However, both theoretical analysis and empirical observations show that a model might perform inconsistently on different measures. To bridge this gap, this paper proposes a novel measure named Top-K Pairwise Ranking (TKPR), and a series of analyses show that TKPR is compatible with existing ranking-based measures. In light of this, we further establish an empirical surrogate risk minimization framework for TKPR. On one hand, the proposed framework enjoys convex surrogate losses with the theoretical support of Fisher consistency. On the other hand, we establish a sharp generalization bound for the proposed framework based on a novel technique named data-dependent contraction. Finally, empirical results on benchmark datasets validate the effectiveness of the proposed framework.
Although attention mechanisms have achieved considerable progress in Transformer-based architectures across various Artificial Intelligence (AI) domains, their inner workings remain to be explored. Existing explainable methods have different emphases but are rather one-sided. They primarily analyse the attention mechanisms or gradient-based attribution while neglecting the magnitudes of input feature values or the skip-connection module. Moreover, they inevitably bring spurious noisy pixel attributions unrelated to the model's decision, hindering humans' trust in the spotted visualization result. Hence, we propose an easy-to-implement but effective way to remedy this flaw: Smooth Noise Norm Attention (SNNA). We weigh the attention by the norm of the transformed value vector and guide the label-specific signal with the attention gradient, then randomly sample the input perturbations and average the corresponding gradients to produce noise-free attribution. Instead of evaluating the explanation method on the binary or multi-class classification tasks like in previous works, we explore the more complex multi-label classification scenario in this work, i.e., the driving action prediction task, and trained a model for it specifically. Both qualitative and quantitative evaluation results show the superiority of SNNA compared to other SOTA attention-based explainable methods in generating a clearer visual explanation map and ranking the input pixel importance.
We study open-world multi-label text classification under extremely weak supervision (XWS), where the user only provides a brief description for classification objectives without any labels or ground-truth label space. Similar single-label XWS settings have been explored recently, however, these methods cannot be easily adapted for multi-label. We observe that (1) most documents have a dominant class covering the majority of content and (2) long-tail labels would appear in some documents as a dominant class. Therefore, we first utilize the user description to prompt a large language model (LLM) for dominant keyphrases of a subset of raw documents, and then construct a (initial) label space via clustering. We further apply a zero-shot multi-label classifier to locate the documents with small top predicted scores, so we can revisit their dominant keyphrases for more long-tail labels. We iterate this process to discover a comprehensive label space and construct a multi-label classifier as a novel method, X-MLClass. X-MLClass exhibits a remarkable increase in ground-truth label space coverage on various datasets, for example, a 40% improvement on the AAPD dataset over topic modeling and keyword extraction methods. Moreover, X-MLClass achieves the best end-to-end multi-label classification accuracy.
The extreme multi-label classification~(XMC) task involves learning a classifier that can predict from a large label set the most relevant subset of labels for a data instance. While deep neural networks~(DNNs) have demonstrated remarkable success in XMC problems, the task is still challenging because it must deal with a large number of output labels, which make the DNN training computationally expensive. This paper addresses the issue by exploring the use of random circular vectors, where each vector component is represented as a complex amplitude. In our framework, we can develop an output layer and loss function of DNNs for XMC by representing the final output layer as a fully connected layer that directly predicts a low-dimensional circular vector encoding a set of labels for a data instance. We conducted experiments on synthetic datasets to verify that circular vectors have better label encoding capacity and retrieval ability than normal real-valued vectors. Then, we conducted experiments on actual XMC datasets and found that these appealing properties of circular vectors contribute to significant improvements in task performance compared with a previous model using random real-valued vectors, while reducing the size of the output layers by up to 99%.
We introduce the sequence classification problem CIViC Evidence to the field of medical NLP. CIViC Evidence denotes the multi-label classification problem of assigning labels of clinical evidence to abstracts of scientific papers which have examined various combinations of genomic variants, cancer types, and treatment approaches. We approach CIViC Evidence using different language models: We fine-tune pretrained checkpoints of BERT and RoBERTa on the CIViC Evidence dataset and challenge their performance with models of the same architecture which have been pretrained on domain-specific text. In this context, we find that BiomedBERT and BioLinkBERT can outperform BERT on CIViC Evidence (+0.8% and +0.9% absolute improvement in class-support weighted F1 score). All transformer-based models show a clear performance edge when compared to a logistic regression trained on bigram tf-idf scores (+1.5 - 2.7% improved F1 score). We compare the aforementioned BERT-like models to OpenAI's GPT-4 in a few-shot setting (on a small subset of our original test dataset), demonstrating that, without additional prompt-engineering or fine-tuning, GPT-4 performs worse on CIViC Evidence than our six fine-tuned models (66.1% weighted F1 score compared to 71.8% for the best fine-tuned model). However, performance gets reasonably close to the benchmark of a logistic regression model trained on bigram tf-idf scores (67.7% weighted F1 score).
This article investigates how well deep learning models can identify web registers -- text varieties such as news reports and discussion forums -- across 16 languages. We introduce the Multilingual CORE corpora, which contain 72,504 documents annotated with a hierarchical taxonomy of 25 registers designed to cover the entire open web. Our multilingual models achieve state-of-the-art results (79% F1 score) using multi-label classification. This performance matches or exceeds previous studies that used simpler classification schemes, showing that models can perform well even with a complex register scheme at a massively multilingual scale. However, we observe a consistent performance ceiling around 77-80% F1 score across all models and configurations. When we remove documents with uncertain labels through data pruning, performance increases to over 90% F1, suggesting that this ceiling stems from inherent ambiguity in web registers rather than model limitations. Analysis of hybrid documents -- texts combining multiple registers -- reveals that the main challenge is not in classifying hybrids themselves, but in distinguishing between hybrid and non-hybrid documents. Multilingual models consistently outperform monolingual ones, particularly helping languages with limited training data. While zero-shot performance drops by an average of 7% on unseen languages, this decrease varies substantially between languages (from 3% to 20%), indicating that while registers share many features across languages, they also maintain language-specific characteristics.
Cross-silo federated learning (FL) enables decentralized organizations to collaboratively train models while preserving data privacy and has made significant progress in medical image classification. One common assumption is task homogeneity where each client has access to all classes during training. However, in clinical practice, given a multi-label classification task, constrained by the level of medical knowledge and the prevalence of diseases, each institution may diagnose only partial categories, resulting in task heterogeneity. How to pursue effective multi-label medical image classification under task heterogeneity is under-explored. In this paper, we first formulate such a realistic label missing setting in the multi-label FL domain and propose a two-stage method FedMLP to combat class missing from two aspects: pseudo label tagging and global knowledge learning. The former utilizes a warmed-up model to generate class prototypes and select samples with high confidence to supplement missing labels, while the latter uses a global model as a teacher for consistency regularization to prevent forgetting missing class knowledge. Experiments on two publicly-available medical datasets validate the superiority of FedMLP against the state-of-the-art both federated semi-supervised and noisy label learning approaches under task heterogeneity. Code is available at https://github.com/szbonaldo/FedMLP.
Vision transformers combined with self-supervised learning have enabled the development of models which scale across large datasets for several downstream tasks like classification, segmentation and detection. The low-shot learning capability of these models, across several low-shot downstream tasks, has been largely under explored. We perform a system level study of different self supervised pretext tasks, namely contrastive learning, clustering, and masked image modelling for their low-shot capabilities by comparing the pretrained models. In addition we also study the effects of collapse avoidance methods, namely centring, ME-MAX, sinkhorn, on these downstream tasks. Based on our detailed analysis, we introduce a framework involving both mask image modelling and clustering as pretext tasks, which performs better across all low-shot downstream tasks, including multi-class classification, multi-label classification and semantic segmentation. Furthermore, when testing the model on full scale datasets, we show performance gains in multi-class classification, multi-label classification and semantic segmentation.
In the field of crisis/disaster informatics, social media is increasingly being used for improving situational awareness to inform response and relief efforts. Efficient and accurate text classification tools have been a focal area of investigation in crisis informatics. However, current methods mostly rely on single-label text classification models, which fails to capture different insights embedded in dynamic and multifaceted disaster-related social media data. This study introduces a novel approach to disaster text classification by enhancing a pre-trained Large Language Model (LLM) through instruction fine-tuning targeted for multi-label classification of disaster-related tweets. Our methodology involves creating a comprehensive instruction dataset from disaster-related tweets, which is then used to fine-tune an open-source LLM, thereby embedding it with disaster-specific knowledge. This fine-tuned model can classify multiple aspects of disaster-related information simultaneously, such as the type of event, informativeness, and involvement of human aid, significantly improving the utility of social media data for situational awareness in disasters. The results demonstrate that this approach enhances the categorization of critical information from social media posts, thereby facilitating a more effective deployment for situational awareness during emergencies. This research paves the way for more advanced, adaptable, and robust disaster management tools, leveraging the capabilities of LLMs to improve real-time situational awareness and response strategies in disaster scenarios.
In Fair AI literature, the practice of maliciously creating unfair models that nevertheless satisfy fairness constraints is known as "cherry-picking". A cherry-picking model is a model that makes mistakes on purpose, selecting bad individuals from a minority class instead of better candidates from the same minority. The model literally cherry-picks whom to select to superficially meet the fairness constraints while making minimal changes to the unfair model. This practice has been described as "blatantly unfair" and has a negative impact on already marginalized communities, undermining the intended purpose of fairness measures specifically designed to protect these communities. A common assumption is that cherry-picking arises solely from malicious intent and that models designed only to optimize fairness metrics would avoid this behavior. We show that this is not the case: models optimized to minimize fairness metrics while maximizing performance are often forced to cherry-pick to some degree. In other words, cherry-picking might be an inevitable outcome of the optimization process itself. To demonstrate this, we use tools from fair cake-cutting, a mathematical subfield that studies the problem of fairly dividing a resource, referred to as the "cake," among a number of participants. This concept is connected to supervised multi-label classification: any dataset can be thought of as a cake that needs to be distributed among different labels, and the model is the function that divides the cake. We adapt these classical results for machine learning and demonstrate how this connection can be prolifically used for fairness and classification in general.
Traditional supervised learning heavily relies on human-annotated datasets, especially in data-hungry neural approaches. However, various tasks, especially multi-label tasks like document-level relation extraction, pose challenges in fully manual annotation due to the specific domain knowledge and large class sets. Therefore, we address the multi-label positive-unlabelled learning (MLPUL) problem, where only a subset of positive classes is annotated. We propose Mixture Learner for Partially Annotated Classification (MLPAC), an RL-based framework combining the exploration ability of reinforcement learning and the exploitation ability of supervised learning. Experimental results across various tasks, including document-level relation extraction, multi-label image classification, and binary PU learning, demonstrate the generalization and effectiveness of our framework.
Multi-label classification is an essential task utilized in a wide variety of real-world applications. Multi-label zero-shot learning is a method for classifying images into multiple unseen categories for which no training data is available, while in general zero-shot situations, the test set may include observed classes. The CLIP-Decoder is a novel method based on the state-of-the-art ML-Decoder attention-based head. We introduce multi-modal representation learning in CLIP-Decoder, utilizing the text encoder to extract text features and the image encoder for image feature extraction. Furthermore, we minimize semantic mismatch by aligning image and word embeddings in the same dimension and comparing their respective representations using a combined loss, which comprises classification loss and CLIP loss. This strategy outperforms other methods and we achieve cutting-edge results on zero-shot multilabel classification tasks using CLIP-Decoder. Our method achieves an absolute increase of 3.9% in performance compared to existing methods for zero-shot learning multi-label classification tasks. Additionally, in the generalized zero-shot learning multi-label classification task, our method shows an impressive increase of almost 2.3%.
We consider sequential maximization of performance metrics that are general functions of a confusion matrix of a classifier (such as precision, F-measure, or G-mean). Such metrics are, in general, non-decomposable over individual instances, making their optimization very challenging. While they have been extensively studied under different frameworks in the batch setting, their analysis in the online learning regime is very limited, with only a few distinguished exceptions. In this paper, we introduce and analyze a general online algorithm that can be used in a straightforward way with a variety of complex performance metrics in binary, multi-class, and multi-label classification problems. The algorithm's update and prediction rules are appealingly simple and computationally efficient without the need to store any past data. We show the algorithm attains $\mathcal{O}(\frac{\ln n}{n})$ regret for concave and smooth metrics and verify the efficiency of the proposed algorithm in empirical studies.
Online communities such as drug-related subreddits serve as safe spaces for people who use drugs (PWUD), fostering discussions on substance use experiences, harm reduction, and addiction recovery. Users' shared narratives on these forums provide insights into the likelihood of developing a substance use disorder (SUD) and recovery potential. Our study aims to develop a multi-level, multi-label classification model to analyze online user-generated texts about substance use experiences. For this purpose, we first introduce a novel taxonomy to assess the nature of posts, including their intended connections (Inquisition or Disclosure), subjects (e.g., Recovery, Dependency), and specific objectives (e.g., Relapse, Quality, Safety). Using various multi-label classification algorithms on a set of annotated data, we show that GPT-4, when prompted with instructions, definitions, and examples, outperformed all other models. We apply this model to label an additional 1,000 posts and analyze the categories of linguistic expression used within posts in each class. Our analysis shows that topics such as Safety, Combination of Substances, and Mental Health see more disclosure, while discussions about physiological Effects focus on harm reduction. Our work enriches the understanding of PWUD's experiences and informs the broader knowledge base on SUD and drug use.
Graph Neural Networks (GNNs) have achieved state-of-the-art results in node classification tasks. However, most improvements are in multi-class classification, with less focus on the cases where each node could have multiple labels. The first challenge in studying multi-label node classification is the scarcity of publicly available datasets. To address this, we collected and released three real-world biological datasets and developed a multi-label graph generator with tunable properties. We also argue that traditional notions of homophily and heterophily do not apply well to multi-label scenarios. Therefore, we define homophily and Cross-Class Neighborhood Similarity for multi-label classification and investigate $9$ collected multi-label datasets. Lastly, we conducted a large-scale comparative study with $8$ methods across nine datasets to evaluate current progress in multi-label node classification. We release our code at \url{https://github.com/Tianqi-py/MLGNC}.
Language Models (LMs) are being proposed for mental health applications where the heightened risk of adverse outcomes means predictive performance may not be a sufficient litmus test of a model's utility in clinical practice. A model that can be trusted for practice should have a correspondence between explanation and clinical determination, yet no prior research has examined the attention fidelity of these models and their effect on ground truth explanations. We introduce an evaluation design that focuses on the robustness and explainability of LMs in identifying Wellness Dimensions (WDs). We focus on two existing mental health and well-being datasets: (a) Multi-label Classification-based MultiWD, and (b) WellXplain for evaluating attention mechanism veracity against expert-labeled explanations. The labels are based on Halbert Dunn's theory of wellness, which gives grounding to our evaluation. We reveal four surprising results about LMs/LLMs: (1) Despite their human-like capabilities, GPT-3.5/4 lag behind RoBERTa, and MedAlpaca, a fine-tuned LLM on WellXplain fails to deliver any remarkable improvements in performance or explanations. (2) Re-examining LMs' predictions based on a confidence-oriented loss function reveals a significant performance drop. (3) Across all LMs/LLMs, the alignment between attention and explanations remains low, with LLMs scoring a dismal 0.0. (4) Most mental health-specific LMs/LLMs overlook domain-specific knowledge and undervalue explanations, causing these discrepancies. This study highlights the need for further research into their consistency and explanations in mental health and well-being.
The automated classification of stuttered speech has significant implications for timely assessments providing assistance to speech language pathologists. Despite notable advancements in the field, the cases in which multiple disfluencies occur in speech require attention. We have taken a progressive approach to fill this gap by classifying multi-stuttered speech more efficiently. The problem has been addressed by firstly curating a dataset of multi-stuttered disfluencies from open source dataset SEP-28k audio clips. Secondly, employing Whisper, a state-of-the-art speech recognition model has been leveraged by using its encoder and taking the problem as multi label classification. Thirdly, using a 6 encoder layer Whisper and experimenting with various layer freezing strategies, a computationally efficient configuration of the model was identified. The proposed configuration achieved micro, macro, and weighted F1-scores of 0.88, 0.85, and 0.87, correspondingly on an external test dataset i.e. Fluency-Bank. In addition, through layer freezing strategies, we were able to achieve the aforementioned results by fine-tuning a single encoder layer, consequently, reducing the model's trainable parameters from 20.27 million to 3.29 million. This research study unveils the contribution of the last encoder layer in the identification of disfluencies in stuttered speech. Consequently, it has led to a computationally efficient approach, 83.7% less parameters to train, making the proposed approach more adaptable for various dialects and languages.