TEXT CLASSIFIERS

Label Sleuth: From Unlabeled Text to a Classifier in a Few Hours (2022.emnlp-demos)

Challenge:	Weakly-supervised text classification aims to induce text classifiers from only a handful of user-provided seed words.
Approach:	They propose to use user-provided seed words to induce text classifiers using only a handful of carefully chosen seed words.
Outcome:	The proposed method outperforms baseline model using only category name seed words and achieves comparable performance as a counterpart using expert-annotated seed words.

Copied to clipboard

Eyal Shnarch, Alon Halfon, Ariel Gera, Marina Danilevsky, Yannis Katsis, Leshem Choshen, Martin Santillan Cooper, Dina Epelboim, Zheng Zhang, Dakuo Wang

Text Classification

Challenge:	Label Sleuth is an open source system for labeling and creating text classifiers which does not require coding skills nor machine learning knowledge.
Approach:	Label Sleuth is an open source system for labeling and creating text classifiers which does not require coding skills nor machine learning knowledge.
Outcome:	Label Sleuth is an open source system for labeling and creating text classifiers.

Metric Learning for Dynamic Text Classification (D19-61)

Copied to clipboard

Jeremy Wohlwend, Ethan R. Elenberg, Sam Altschul, Shawn Henry, Tao Lei

Unsupervised Reinforcement Adaptation for Class-Imbalanced Text Classification (2022.starsem-1)

Challenge:	Traditional text classifiers are limited to predicting over a fixed set of labels, but real-world applications require dynamic classification.
Approach:	They propose to replace the traditional fixed-size output layer with a learned metric space . they propose to add or remove support points in the metric and fine-tune the resulting metric .
Outcome:	The proposed method is robust to changes in the label space and improves performance in low data regime.

Copied to clipboard

Yuexin Wu, Xiaolei Huang

BFClass: A Backdoor-free Text Classification Framework (2021.findings-emnlp)

Challenge:	Existing models for class imbalanced labels learn domain-invariant representations across domains and evaluate primarily on class-balanced data.
Approach:	They propose an unsupervised domain adaptation approach that leverages feature variants and imbalanced labels across domains to learn robust representations.
Outcome:	The proposed method can learn robust domain-invariant representations and adapt classifiers on imbalanced classes over domains.

Copied to clipboard

Zichao Li, Dheeraj Mekala, Chengyu Dong, Jingbo Shang

R&R: Metric-guided Adversarial Sentence Generation (2022.findings-aacl)

Challenge:	Various trigger design strategies have been explored to attack text classifiers, however, defending such attacks remains an open problem.
Approach:	They propose a backdoor-free training framework that poisons a subset of training data by injecting trigger patterns and setting their labels as the target labels.
Outcome:	The proposed framework can detect all the triggers, remove 95% of poisoned training samples with very limited false alarms, and achieve almost the same performance as the models trained on benign training data.

Copied to clipboard

Lei Xu, Alfredo Cuesta-Infante, Laure Berti-Equille, Kalyan Veeramachaneni

Easy Adaptation to Mitigate Gender Bias in Multilingual Text Classification (2022.naacl-main)

Challenge:	Existing methods prioritize misclassification by maximizing each perturbation’s effectiveness at misleading a text classifier.
Approach:	They propose a rewrite and rollback framework for adversarial attack that optimizes a critique score which combines fluency, similarity, and misclassification metrics.
Outcome:	The proposed framework outperforms current state-of-the-art in attack success rate by +16.2%, +12.8%, and +14.0% on the classifiers respectively.

Copied to clipboard

Xiaolei Huang

Multilinguality

Challenge:	Existing approaches to mitigate demographic biases evaluate on monolingual data, however, multilingual data has not been examined.
Approach:	They propose a standard domain adaptation model to reduce gender bias in multilingual contexts.
Outcome:	The proposed model reduces gender bias and improves on two text classification tasks with three fair-aware baselines.

Integrated Directional Gradients: Feature Interaction Attribution for Neural NLP Models (2021.acl-long)

Copied to clipboard

Sandipan Sikdar, Parantapa Bhattacharya, Kieran Heese

Semantic Text Processing

Challenge:	Existing methods for attribution of importance to features borrowed from cooperative game theory . success of Deep Neural Networks has led to their ability to learn from complex higher order interactions from raw features.
Approach:	They propose a method for attributing importance scores to groups of features . they propose axioms that any intuitive feature group attribution method should satisfy .
Outcome:	The proposed method captures the importance of features in a linguistic model using negations and conjunctions.

The Promises and Pitfalls of LLM Annotations in Dataset Labeling: a Case Study on Media Bias Detection (2025.findings-naacl)

Copied to clipboard

Tomáš Horych, Christoph Mandl, Terry Ruas, Andre Greiner-Petter, Bela Gipp, Akiko Aizawa, Timo Spinde

Exploiting Class Probabilities for Black-box Sentence-level Attacks (2024.findings-eacl)

Challenge:	Recent research suggests using Large Language Models (LLMs) to automate the annotation process, reducing these costs while maintaining data quality.
Approach:	They propose to use Large Language Models to automate annotation process and train classifiers on large datasets.
Outcome:	The proposed model outperforms all of the annotator LLMs on two media bias benchmark datasets (BABE and BASIL) while maintaining data quality.

Copied to clipboard

Raha Moraffah, Huan Liu

Smaller Text Classifiers with Discriminative Cluster Embeddings (N18-2)

Challenge:	Existing sentences use either no feedback or only the class labels for sentence-level attacks.
Approach:	They propose an algorithm that uses class probabilities for black-box sentence-level attacks and investigate the effectiveness of using class probabilties on the attack’s success.
Outcome:	The proposed algorithm is evaluated against baselines and classifiers and compares with the existing models to determine whether it is worthy or practical to use class probabilities for black-box sentence-level attacks.

Copied to clipboard

Mingda Chen, Kevin Gimpel

Efficient, Uncertainty-based Moderation of Neural Networks Text Classifiers (2022.findings-acl)

Challenge:	Word embeddings dominate overall model sizes in neural methods for natural language processing, especially when large vocabularies and high dimensions are used.
Approach:	They propose a Gumbel-Softmax distribution to maximize over the latent clustering while minimizing the task loss.
Outcome:	The proposed method minimizes the task loss while maximizing over the latent clustering while remaining parameter-efficient.

Copied to clipboard

Jakob Smedegaard Andersen, Walid Maalej

Towards Adversarially Robust Text Classifiers by Learning to Reweight Clean Examples (2022.findings-acl)

Challenge:	A series of benchmarking experiments based on three different datasets and three state-of-the-art classifiers show that our framework can improve the classification F1-scores by 5.1 to 11.2% (up to approx. 98 to 99%)
Approach:	They propose a semi-automated approach that passes unconfident, probably incorrect classifications to human moderators to minimize the workload.
Outcome:	The proposed approach can improve the classification F1-scores by 5.1 to 11.2% (up to approx. 98 to 99%) while reducing the moderation load up to 73.3% compared to a random moderation.

Copied to clipboard

Jianhan Xu, Cenyuan Zhang, Xiaoqing Zheng, Linyang Li, Cho-Jui Hsieh, Kai-Wei Chang, Xuanjing Huang

ChatGPT as an Attack Tool: Stealthy Textual Backdoor Attack via Blackbox Generative Model Trigger (2024.naacl-long)

Challenge:	Existing defense methods improve the adversarial robustness by making models adapt to training set augmented with some adversarials.
Approach:	They propose to introduce a reweighting mechanism to calibrate the training distribution to obtain robust models.
Outcome:	The proposed method minimizes the loss of validation set mixed with clean examples and adversarial ones in an online learning manner.

Copied to clipboard

Jiazhao Li, Yijin Yang, Zhuofeng Wu, V.G.Vinod Vydiswaran, Chaowei Xiao

Investigating the Working of Text Classifiers (C18-1)

Challenge:	Textual backdoor attacks are increasingly challenging to detect due to the use of advanced generative models such as GPT-4.
Approach:	They propose a framework that harnesses advanced generative models to execute stealthier backdoor attacks on text classifiers.
Outcome:	The proposed framework achieves state-of-the-art attack success rate of 97.35% over four sentiment classification tasks and four human cognition stealthiness tests.

Copied to clipboard

Devendra Sachan, Manzil Zaheer, Ruslan Salakhutdinov

Necessity and Sufficiency for Explaining Text Classifiers: A Case Study in Hate Speech Detection (2022.naacl-main)

Challenge:	Text classification is one of the most widely studied tasks in natural language processing.
Approach:	They propose to use large multilayer neural network models to compose meaning of sentences . they propose to disincentivize focusing on key lexicons to improve classification accuracy .
Outcome:	The proposed models learn to compose the meaning of the sentences or focus on key lexicons for classifying the document.

Copied to clipboard

Esma Balkir, Isar Nejadgholi, Kathleen Fraser, Svetlana Kiritchenko

Bag-of-Words vs. Graph vs. Sequence in Text Classification: Questioning the Necessity of Text-Graphs and the Surprising Strength of a Wide MLP (2022.acl-long)

Challenge:	XAI features usually provide a single importance score for each token, but feature attribution methods provide two complementary and theoretically-grounded scores for each utterance.
Approach:	They propose a feature attribution method that generates explicit perturbations of the input text, allowing the importance scores themselves to be explainable.
Outcome:	The proposed method explain the predictions of hate speech detection models on a set of curated examples from a test suite.

Copied to clipboard

Lukas Galke, Ansgar Scherp

Narrative Style and the Spread of Health Misinformation on Twitter (2023.findings-emnlp)

Challenge:	Existing graph-based text classification methods are based on a Bag-of-Words (BoW) .
Approach:	They propose to use a Bag-of-Words-based multi-layer perceptron model to perform inductive text classification using a sequence-based BERT and a lightweight DistilBERT model to fine-tune their results.
Outcome:	The proposed model outperforms the current graph-based models and questions the importance of synthetic graphs in modern text classification methods.

Copied to clipboard

Achyutarama Ganti, Eslam Ali Hassan Hussein, Steven Wilson, Zexin Ma, Xinyan Zhao

Interpretable Neural Predictions with Differentiable Binary Variables (P19-1)

Challenge:	Using a narrative style is an effective way to communicate health information on and off social media platforms.
Approach:	They annotate health misinformation tweets and classify them into narrative and non-narrative . they then use supervised fine-tuning and in-context learning to detect narratives .
Outcome:	The proposed model analyzes health misinformation tweets and finds that narrative use is linked to increased tweet engagement and can lead to increased misinformation use.

Copied to clipboard

Jasmijn Bastings, Wilker Aziz, Ivan Titov

Explainability & Interpretability in NLP

Challenge:	Neural networks are bringing incredible performance gains on text classification tasks, but they also require interpretability.
Approach:	They propose a latent model that selects a rationale and a classifier that learns from the words in the rationale alone.
Outcome:	The proposed model can predict expected value of penalties without REINFORCE and can be directly optimised towards a pre-specified text selection rate.

Universal Adversarial Attacks with Natural Triggers for Text Classification (2021.naacl-main)

Copied to clipboard

Liwei Song, Xinwei Yu, Hsuan-Tung Peng, Karthik Narasimhan

Identifying Spurious Correlations for Robust Text Classification (2020.findings-emnlp)

Challenge:	Recent work has demonstrated the vulnerability of modern text classifiers to universal adversarial attacks, which are input-agnostic sequences of words added to text processed by classifier.
Approach:	They propose a gradient-based search that aims to maximize the downstream classifier’s prediction loss by using an adversarially regularized autoencoder to generate triggers and propose heuristics to spot such attacks.
Outcome:	The proposed algorithms reduce model accuracy while being less identifiable than prior models as per automatic detection metrics and human-subject studies.

Copied to clipboard

Zhao Wang, Aron Culotta

Hype or not? Formalizing Automatic Promotional Language Detection in Biomedical Research (2026.eacl-long)

Challenge:	Text classifiers often rely on spurious correlations to predict positive reviews . term Spielberg does not cause the review to be positive, so it does not affect the classification accuracy.
Approach:	They propose a method to distinguish spurious and genuine correlations in text classification using treatment effect estimators.
Outcome:	The proposed method works well even with limited training examples and is possible to transport the word classifier to new domains.

Copied to clipboard

Bojan Batalo, Erica K. Shimomoto, Dipesh Satav, Neil Millar

Challenge:	Promotional language is a term used to undermine objective evaluation of evidence, impede research development, and erode trust in science.
Approach:	They propose formalized guidelines for identifying hype language and apply them to annotate a portion of the National Institutes of Health grant application corpus.
Outcome:	The proposed guidelines can help humans reliably annotate candidate hype adjectives and train machine learning models yield promising results.

Counterfactual Inference for Text Classification Debiasing (2021.acl-long)

Copied to clipboard

Chen Qian, Fuli Feng, Lijie Wen, Chunping Ma, Pengjun Xie

Topics to Avoid: Demoting Latent Confounds in Text Classification (D19-1)

Challenge:	Existing methods to capture unintended dataset biases are expensive and require elaborate balancing strategies.
Approach:	They propose a model-agnostic text classification debiasing framework which can effectively avoid employing data manipulations or designing balancing mechanisms.
Outcome:	The proposed framework can effectively avoid data manipulations or designing balancing mechanisms to capture unintended dataset biases.

Copied to clipboard

Sachin Kumar, Shuly Wintner, Noah A. Smith, Yulia Tsvetkov

Text Classification

Challenge:	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.
Approach:	They propose a method that represents latent topical confounds and a model which “unlearns” confounding features by predicting both the label of the input text and the confound.
Outcome:	The proposed model generalizes better and learns features indicative of the writing style rather than the content.

Explanation Regularisation through the Lens of Attributions (2025.coling-main)

Copied to clipboard

Pedro Ferreira, Ivan Titov, Wilker Aziz

Explainability & Interpretability in NLP

Challenge:	Explanation regularisation (ER) is a method to guide text classifiers to form their predictions relying on tokens that humans consider plausible.
Approach:	They introduce an auxiliary explanation loss to measure how well an input attribution technique's output agrees with human-annotated rationales.
Outcome:	The proposed model improves classification performance in out-of-domain (OOD) settings by relying on tokens humans consider plausible.

Speak No Evil, Just Prompt: Low-resource Multilingual Toxic Speech Detection with Audio Language Model (2026.findings-acl)

Copied to clipboard

Mingzi Zuo, Lei Zhang, Hailiang Sun, Shengzhi Huo, Changyu Dong, Xin Wang, Bo Wang, Hao Liu

Cross Domain Classification of Education Talk Turns (2025.coling-main)

Challenge:	Existing methods for toxic speech detection rely on high-resource languages and lack acoustic cues.
Approach:	They propose a prompt-based adaptation framework that performs end-to-end toxicity detection without ASR.
Outcome:	The proposed framework achieves a micro-averaged ROC-AUC of 98.07% on polySpeechTox . it is based on a frozen audio language model and can perform end-to-end toxicity detection without ASR .

Copied to clipboard

Achyutarama R. Ganti, Steven R. Wilson, Geoffrey Louie Wing-Yue

Text Classification

Challenge:	Prior research has focused on the annotation of conversational talk-turns within the classroom, offering a statistical analysis of the various types of discourse prevalent in these environments.
Approach:	They examine the generalizability and transferability of text classifiers trained to predict classroom discourse across educational domains by accompanying each talk turn with dialog-level context.
Outcome:	The proposed models exhibit high generalizability when training and test datasets originate from the same or similar domains.

Word-Level Uncertainty Estimation for Black-Box Text Classifiers using RNNs (2020.coling-main)

Copied to clipboard

Jakob Smedegaard Andersen, Tom Schöner, Walid Maalej

Challenges of Using Text Classifiers for Causal Inference (D18-1)

Challenge:	Neural Networks are not interpretable, since they provide no information about why particular decisions were made.
Approach:	They propose to decompose and visualize uncertainty of text classifiers at the level of words to provide detailed explanations of uncertainties.
Outcome:	The proposed approach decomposes and visualizes uncertainty of text classifiers at the level of words and enables a deeper understanding of unreliable model behaviours.

Copied to clipboard

Zach Wood-Doughty, Ilya Shpitser, Mark Dredze

Few-Shot Learning with Siamese Networks and Label Tuning (2022.acl-long)

Challenge:	a number of scientific analyses focus on low-dimensional structured data, but text classifiers can be used to produce structured variables.
Approach:	They propose to use text classifiers to conduct causal analyses on simulated and Yelp data.
Outcome:	The proposed method can be used on simulated and Yelp data.

Copied to clipboard

Thomas Müller, Guillermo Pérez-Torró, Marc Franco-Salvador

Low-Resource NLP

Challenge:	Recent studies have shown that few-shot text classification is a poor solution for training data-intensive tasks.
Approach:	They propose a method that embeds texts and labels into classifiers with proper pre-training.
Outcome:	The proposed approach reduces inference cost by increasing the number of labels and embeddings.

Textual Data Augmentation for Efficient Active Learning on Tiny Datasets (2020.emnlp-main)

Copied to clipboard

Husam Quteineh, Spyridon Samothrakis, Richard Sutcliffe

Incorporating Priors with Feature Attribution on Text Classification (P19-1)

Challenge:	Existing active learning approaches for textual data are limited due to the complexity of language.
Approach:	They propose an approach where guided outputs of a language generation model can be enhanced through an active learning process.
Outcome:	The proposed approach achieves performance increases of 3% and 5% on TREC-6 and SST-2 datasets compared with NGDG, which does not optimize for a reward function.

Copied to clipboard

Frederick Liu, Besim Avci

AIA-BDE: A Corpus of FAQs in Portuguese and their Variations (2020.lrec-1)

Challenge:	Feature attribution methods are used to help users interpret complex models.
Approach:	They propose a feature attribution method that integrates feature attributed features into the objective function to allow machine learning practitioners to incorporate priors in model building.
Outcome:	The proposed method reduces undesired model biases without a tradeoff on the original task and improves classifier performance in scarce data setting.

Copied to clipboard

Hugo Gonçalo Oliveira, João Ferreira, José Santos, Pedro Fialho, Ricardo Rodrigues, Luisa Coheur, Ana Alves

Don’t Retrain, Just Rewrite: Countering Adversarial Perturbations by Rewriting Text (2023.acl-long)

Challenge:	a corpus of 380 domain-oriented FAQs in Portuguese is presented . paraphrases or entailed questions are created manually, by humans, or automatically, with Google Translate.
Approach:	They present a corpus of 380 domain-oriented FAQs in Portuguese and their variations, i.e., paraphrases or entailed questions, created manually, by humans, or automatically, with Google Translate.
Outcome:	The proposed system outperforms other systems in the domain of question-answering . it performs well when matching variations with their original questions .

Copied to clipboard

Ashim Gupta, Carter Blum, Temma Choji, Yingjie Fei, Shalin Shah, Alakananda Vempala, Vivek Srikumar