Generative text classifiers are most vulnerable to membership inference attacks.
problem Privacy threat from Membership Inference Attacks on generative text classifiers.
method Comprehensive empirical evaluation of generative, discriminative, and pseudo-generative classifiers across various datasets.
result Generative classifiers explicitly modeling P(X,Y) are most vulnerable to membership leakage. We propose a generic and interpretable learning framework for building robust text classification model that achieves accuracy comparable to full models under test-time budget constraints. Our approach learns a selector to identify words that are relevant to the prediction tasks and passes them to the classifier for pr…
Classifier detects LLM-generated text with guarantees.
problem Detecting fake LLM-generated text to prevent misuse.
method Trained classifier without auxiliary info, distinguishing human and LLM text.
result Achieves higher accuracy than existing detectors with type-I error control.
PySS3 simplifies access to SS3's text classification and visualization.
problem Limited availability of an open-source SS3 implementation.
method Developed PySS3, an open-source Python package implementing SS3.
result PySS3 enables robust, explainable, and trusty text classification.
This research discovers model architecture and training dataset characteristics through strategic input probing.
problem Discovering model architecture and training dataset characteristics in black box models.
method Structured input probes and model outputs are used to train a deep classifier for image and text classification.
result The approach successfully distinguishes between different image and text datasets and architectures.
Adversarial examples are carefully constructed modifications to an input that completely change the output of a classifier but are imperceptible to humans. Despite these successful attacks for continuous data (such as image and audio samples), generating adversarial examples for discrete structures such as text has pro…
Paper evaluates using app images for classification, improving accuracy.
problem Improving app classification accuracy when text descriptions are missing or inadequate.
method Used OCR, pic2vec, captionbot.ai, and object detection to convert images into text or vectors for classification.
result Improved classification accuracy of 96% for some app categories when images are added.
Paper develops heavy-tailed embeddings for better text classification and augmentation.
problem Improving text classification, especially for extreme values.
method Develops heavy-tailed embeddings using multivariate extreme value theory and introduces a scale-invariant classifier.
result The classifier outperforms baselines and generates meaningful augmented text.
New method explains high-dimensional text classifiers.
problem Limited explainability tools for high-dimensional inputs and neural networks.
method Theoretical high-dimensional properties in neural networks.
result Improved explainability for neural network classifiers.
Universal perturbations misclassify text with high accuracy.
problem Vulnerability of text classifiers to small perturbations.
method Algorithm to compute universal adversarial perturbations.
result Deep neural networks are highly vulnerable to universal adversarial perturbations.
A great variety of text tasks such as topic or spam identification, user profiling, and sentiment analysis can be posed as a supervised learning problem and tackle using a text classifier. A text classifier consists of several subprocesses, some of them are general enough to be applied to any supervised learning proble…
Classifies manifolds with specific spinors and constructs parallel spinors.
problem Classifying manifolds with generalized Killing spinors.
method Explicitly constructs parallel spinors and considers modified Dirac currents.
result Classifies Riemannian spin^c manifolds with type I and II imaginary generalized Killing spinors.
Despite impressive performance on many text classification tasks, deep neural networks tend to learn frequent superficial patterns that are specific to the training data and do not always generalize well. In this work, we observe this limitation with respect to the task of native language identification. We find that s…
Study shows text-based news veracity models don't generalize across U.S. and U.K.
problem Generalizability of text-based news veracity detection models across countries.
method Testing news veracity models on U.S. and U.K. news data.
result Text-based classifiers perform poorly when trained on one country's news data and tested on another.
Cross-lingual Text Classification (CLC) consists of automatically classifying, according to a common set C of classes, documents each written in one of a set of languages L, and doing so more accurately than when naively classifying each document via its corresponding language-specific classifier. In order to obtain an…
The paper classifies nilmanifolds with specific SL(3,C) structures.
problem Classifying nilmanifolds with invariant mean convex or tamed SL(3,C) structures.
method Invariants and classification of structures on nilmanifolds.
result Classification of nilmanifolds with invariant mean convex closed SL(3,C) structures.
Guided Flows enhance sample quality in conditional image generation and text-to-speech.
problem Improving sample quality in conditional generative models.
method Integrating classifier-free guidance into Flow Matching (FM) models for Continuous Normalizing Flows (CNFs).
result Guided Flows significantly improve sample quality in conditional image generation and text-to-speech synthesis.
Anchors explains text classifiers by highlighting key words.
problem Interpreting machine learning models, especially for text classifiers.
method Formalizes Anchors' algorithm and analyzes its behavior on linear text classifiers.
result Anchors produces meaningful results on linear text classifiers.
Homomorphisms between pure mapping class groups are classified for certain genus surfaces.
problem Classifying homomorphisms between pure mapping class groups for specific genus surfaces.
method Proving every multitwist-preserving map is induced by a multi-embedding, then applying to classify homomorphisms.
result All homomorphisms between pure mapping class groups for g≥4 and g′≤6⋅2g−4 are classified. New families of Lie groups with special foliations discovered.
problem Characterizing left-invariant foliations on semi-Riemannian Lie groups.
method Classifying foliations generated by specific subgroups.
result Constructing new families of Lie groups with conformal minimal foliations.
Ontology learning is a critical task in industry, dealing with identifying and extracting concepts captured in text data such that these concepts can be used in different tasks, e.g. information retrieval. Ontology learning is non-trivial due to several reasons with limited amount of prior research work that automatica…
Text-based analysis methods allow to reveal privacy relevant author attributes such as gender, age and identify of the text's author. Such methods can compromise the privacy of an anonymous author even when the author tries to remove privacy sensitive content. In this paper, we propose an automatic method, called Adver…
Paper generates diverse, readable adversarial texts from scratch.
problem Text classification models are easily fooled by adversarial examples.
method Trained a conditional variational autoencoder (VAE) with adversarial loss and utilized GANs to generate consistent adversarial texts.
result Successfully generates adversarial texts with higher success rate and acceptable quality.
Efficiently distills pretrained text-to-image models without real data, improving FID and CLIP scores.
problem Slow iterative refinement process of diffusion-based text-to-image models.
method Guided Score identity Distillation with Long and Short Classifier-Free Guidance.
result Achieves state-of-the-art FID performance with competitive CLIP score.
Classifies and constructs intertwining differential operators between line and vector bundles over real projective space.
problem Classifying and constructing intertwining differential operators between line and vector bundles over real projective space.
method F-method for classification and construction of intertwining differential operators.
result Generalizes a classical result of Bol for SL(2,R) and classifies intertwining operators for SL(n,R). In this paper, we study counterfactual fairness in text classification, which asks the question: How would the prediction change if the sensitive attribute referenced in the example were different? Toxicity classifiers demonstrate a counterfactual fairness issue by predicting that "Some people are gay" is toxic while "…
Sentiment analysis (SA) is a task related to understanding people's feelings in written text; the starting point would be to identify the polarity level (positive, neutral or negative) of a given text, moving on to identify emotions or whether a text is humorous or not. This task has been the subject of several researc…
With the widespread use of the internet, the size of the text data increases day by day. Poems can be given as an example of the growing text. In this study, we aim to classify poetry according to poet. Firstly, data set consisting of three different poetry of poets written in English have been constructed. Then, text …
The study classifies homomorphisms from mapping class groups using finite subgroups.
problem Classifying homomorphisms from mapping class groups.
method Using finite subgroups to classify homomorphisms.
result Only finitely many mapping class groups have non-trivial homomorphisms into Homeo(S^n) for any n.
This work uses image generation models to find vision model bugs.
problem Automatically discovering failures in vision models.
method Conditional text-to-image generation and captioning models.
result Demonstrated utility of large-scale generative models to find vision model bugs.
New formula classifies product reviews into higher and lower ratings based on sentiment analysis.
problem Lack of research on using sentiment analysis for classifying text into ratings.
method Redefined sentiment proportions as a triangle structure to derive variables for classifying text into higher and lower ratings.
result Proved a dependence exists between sentiments and ratings.
Graph attention network improves MLTC by capturing label dependencies.
problem Ignoring label dependencies in MLTC tasks.
method Graph attention network model that captures label dependencies.
result The model achieves similar or better performance than state-of-the-art models.
Rule-based classifiers quantify uncertainty using Bernoulli random variables.
problem Quantifying the uncertainty of precision estimates for rule-based text classifiers.
method Treat partitions of sub-strings as Bernoulli random variables, compare means using statistical tests, and combine classifiers using Dempster-Shafer theory.
result The approach can be used to combine binary classifiers into a multi-label classifier.
Traditional classifiers can generate high-quality images comparable to generative models.
problem Separation between classifiers and generators in neural networks.
method Optimizing input gradients to produce images, using mask-based stochastic reconstruction, progressive-resolution technique, and distance metric loss.
result Traditional classifiers can generate high-fidelity images of 256imes256 resolution on ImageNet. Computer generated academic papers have been used to expose a lack of thorough human review at several computer science conferences. We assess the problem of classifying such documents. After identifying and evaluating several quantifiable features of academic papers, we apply methods from machine learning to build a b…
Paper introduces hierarchical softmax for global hierarchical classification tasks.
problem Improving classification accuracy in tasks with class hierarchies.
method Global hierarchical neural networks using hierarchical softmax.
result Hierarchical softmax outperforms regular softmax in multiple datasets.
Decoding strategies often exclude human-like tokens, creating a detectable gap in generated text.
problem Decoding strategies exclude contextually appropriate but statistically rare tokens, creating a detectable gap in generated text.
method Analysis of 1.8 million texts across 8 language models, 5 decoding strategies, and 53 hyperparameter configurations.
result 8-18% of human-selected tokens fall outside typical truncation boundaries, indicating a detectable gap.
As the Portable Document Format (PDF) file format increases in popularity, research in analysing its structure for text extraction and analysis is necessary. Detecting headings can be a crucial component of classifying and extracting meaningful data. This research involves training a supervised learning model to detect…
Improves text-to-SQL models by selecting the best SQL query from beam output.
problem Simplifying database query writing for natural language questions.
method Discriminative re-ranker using BERT fine-tuned classifier.
result Achieved top 4 score on Spider leaderboard.
Recent approaches based on artificial neural networks (ANNs) have shown promising results for short-text classification. However, many short texts occur in sequences (e.g., sentences in a document or utterances in a dialog), and most existing ANN-based systems do not leverage the preceding short texts when classifying …
Study challenges the necessity of data augmentation for improving predictions on imbalanced text datasets.
problem Improving predictions on imbalanced text datasets.
method Comparing classifier cutoff adjustments to data augmentation techniques.
result Classifier cutoff adjustments can produce similar results to data augmentation without the need for additional data.
We prove a homological version of a conjecture about the homotopy type of diffeomorphism spaces of reducible 3-manifolds.
problem Proving a conjecture about the homotopy type of diffeomorphism spaces of reducible 3-manifolds.
method Homological approach to show finitely many nonzero homology groups, each finitely generated.
result BDiff(M, rel ∂) has finitely many nonzero homology groups, each finitely generated, for connected sums of irreducible 3-manifolds with nontrivial and non-spherical boundaries.
We explore two techniques which use color to make sense of statistical text models. One method uses in-text annotations to illustrate a model's view of particular tokens in particular documents. Another uses a high-level, "words-as-pixels" graphic to display an entire corpus. Together, these methods offer both zoomed-i…
New samplers improve compositional generation with diffusion models.
problem Improving compositional generation with diffusion models.
method Score-based interpretation, energy-based parameterization, Metropolis-corrected samplers.
result New samplers enable successful compositional generation across various tasks.
In this project we analysed how much semantic information images carry, and how much value image data can add to sentiment analysis of the text associated with the images. To better understand the contribution from images, we compared models which only made use of image data, models which only made use of text data, an…
Recently, generating adversarial examples has become an important means of measuring robustness of a deep learning model. Adversarial examples help us identify the susceptibilities of the model and further counter those vulnerabilities by applying adversarial training techniques. In natural language domain, small pertu…
The paper classifies actions of a specific group on certain manifolds.
problem Classifying analytic actions of a specific semi-orthogonal group on manifolds.
method Adapting Uchida's construction, the paper explicitly constructs actions on specific manifolds and demonstrates that any action is covered by these.
result Any analytic action of the semi-orthogonal group on a closed, connected manifold is covered by the constructed actions.
New method evaluates text-to-image synthesis for realism, variety, and semantic accuracy.
problem Lack of metrics revealing semantic accuracy in text-to-image synthesis.
method Uses Inception network representations and t-SNE visualization for semantic evaluation.
result Classification accuracy of generated images to real images' visual concepts correlates with semantic accuracy.