FEATURE ENGINEERING

Do Sentence Interactions Matter? Leveraging Sentence Level Representations for Fake News Classification (D19-53)

Challenge:	Existing approaches to the Cloze Test that use feature engineering to achieve high accuracy are ignoring the training set and training a model on the validation set.
Approach:	They propose a fully-neural approach to the Cloze Test using skip-thought embeddings of the stories in a feed-forward network that achieves close to state-of-the-art performance without any feature engineering.
Outcome:	The proposed approach achieves close to state-of-the-art performance on the Cloze Test without any feature engineering.

Copied to clipboard

Vaibhav Vaibhav, Raghuram Mandyam, Eduard Hovy

Ethical NLP

Challenge:	Existing methods to distinguish between trusted and fake news articles lack feature engineering . et al. (2009) define fake news as the one which deliberately exposes real-world individuals, organisations and events to ridicule.
Approach:	They propose a graph neural network-based model which captures sentence interactions within a document.
Outcome:	The proposed model beats baselines and achieves state-of-the-art accuracy on existing datasets.

Neural Math Word Problem Solver with Reinforcement Learning (C18-1)

Copied to clipboard

Danqing Huang, Jing Liu, Chin-Yew Lin, Jian Yin

Reasoning

Challenge:	Existing models for solving math word problems rely on predefined rules or feature engineering.
Approach:	They propose to incorporate copy and alignment mechanism into the sequence-to-sequence model to address two shortcomings . they use model output as a feature and incorporate it into the feature-based model to explore the effectiveness .
Outcome:	The proposed model outperforms the state-of-the-art models on the problem solving task.

Practical Application of Domain Dependent Confidence Measurement for Spoken Language Understanding Systems (N18-3)

Copied to clipboard

Mahnoosh Mehrabani, David Thomson, Benjamin Stern

Transforming Podcast Preview Generation: From Expert Models to LLM-Based Systems (2025.acl-industry)

Challenge:	a confidence score is a scalar quantity that measures the reliability of an automatic system.
Approach:	They propose to use a confidence measure to evaluate the reliability of an SLU system . they build confidence models for three different types of dialogue states .
Outcome:	The proposed model can be used to reject low-confidence SLU results in real-world scenarios.

Copied to clipboard

Winstead Zhu, Ann Clifton, Azin Ghazimatin, Edgar Tanaka, Ward Ronan

PromptFE: Automated Feature Engineering by Prompting (2026.eacl-long)

Challenge:	Podcasts, videos, and other long-form talk content requires significant time investment to assess their relevance.
Approach:	They propose an LLM-based approach for generating podcast episode previews and deploy it at scale, serving hundreds of thousands of podcast previews in a real-world application.
Outcome:	The proposed approach outperforms a baseline built on top of various ML expert models and offers a 4.6% increase in user engagement with preview content and a 5x boost in processing efficiency.

Copied to clipboard

Yufeng Zou, Jean Utke, Diego Klabjan, Han Liu

Leveraging Contextual Information for Effective Entity Salience Detection (2024.findings-naacl)

Challenge:	Existing approaches to feature engineering relied on domain expertise to build features.
Approach:	They propose a framework that leverages large language models to automatically construct features in a string format and generate semantic explanations based on dataset descriptions.
Outcome:	The proposed framework outperforms state-of-the-art methods on real-world datasets.

Copied to clipboard

Rajarshi Bhowmik, Marco Ponza, Atharva Tendle, Anant Gupta, Rebecca Jiang, Xingyu Lu, Qian Zhao, Daniel Preotiuc-Pietro

A System for Diacritizing Four Varieties of Arabic (D19-3)

Challenge:	Prior work on salient entity detection focused on machine learning models that require heavy feature engineering.
Approach:	They propose to fine-tune medium-sized language models with a cross-encoder style architecture to achieve significant performance gains over feature engineering approaches.
Outcome:	The proposed model fine-tunes medium-sized pre-trained language models with a cross-encoder style architecture yields substantial performance gains over feature engineering approaches.

Copied to clipboard

Hamdy Mubarak, Ahmed Abdelali, Kareem Darwish, Mohamed Eldesouki, Younes Samih, Hassan Sajjad

Effort Estimation in Named Entity Tagging Tasks (2020.lrec-1)

Challenge:	Short vowels, aka diacritics, are omitted when writing different varieties of Arabic . diacritization is essential for language learning and text-to-speech applications .
Approach:	They propose a system for recovering diacritics in Arabic without short vowels . they use a character-based sequence-to-sequence deep learning model .
Outcome:	The proposed system beats all previous SOTA systems for Arabic varieties . it uses a character-based sequence-to-sequence deep learning model .

Copied to clipboard

Inês Gomes, Rui Correia, Jorge Ribeiro, João Freitas

Tagging

Challenge:	Named Entity Recognition (NER) requires large amounts of annotated data.
Approach:	They investigate which variables influence the time spent on a named entity annotation task by a human . they found a root mean squared error (RMSE) of 25.68 words per minute with a Nearest Neighbors model .
Outcome:	The proposed model achieves a root mean squared error (RMSE) of 25.68 words per minute with a Nearest Neighbors model.

QUILL: Query Intent with Large Language Models using Retrieval Augmentation and Multi-stage Distillation (2022.emnlp-industry)

Copied to clipboard

Krishna Srinivasan, Karthik Raman, Anupam Samanta, Lingrui Liao, Luca Bertelli, Michael Bendersky

EM-PERSONA: EMotion-assisted Deep Neural Framework for PERSONAlity Subtyping from Suicide Notes (2022.coling-1)

Challenge:	Large Language Models (LLMs) have shown impressive results on a variety of text understanding tasks.
Approach:	They propose a two-stage distillation approach that allows retrieval augmentation to be carried over without the increased compute associated with it.
Outcome:	The proposed approach can carry over the gains of retrieval augmentation without suffering the increased compute typically associated with it.

Copied to clipboard

Soumitra Ghosh, Dhirendra Kumar Maurya, Asif Ekbal, Pushpak Bhattacharyya

Sentiment Analysis

Challenge:	Suicide continues to be one of the significant causes of death worldwide . EMotion-assisted personality subtyping is a novel approach to identify personality traits from suicide notes .
Approach:	They propose to use a PERSONAlity Detection Framework to identify personality traits from suicide notes and annotate them using a benchmark dataset.
Outcome:	The proposed method outperforms baselines on comprehensive evaluation using multiple state-of-the-art systems.

Semi-Supervised Semantic Role Labeling with Cross-View Training (D19-1)

Copied to clipboard

Rui Cai, Mirella Lapata

Responsible & Trustworthy NLP

Challenge:	Recent approaches rely on expensive annotations and are unavailable in low resource scenarios (e.g., rare languages or domains).
Approach:	They propose an end-to-end SRL model which leverages unlabeled data and propose to reduce the annotation effort involved via semi-supervised learning.
Outcome:	The proposed model outperforms the state-of-the-art in English and consistently improves performance in other languages, including Chinese, German, and Spanish.

Mode Effects’ Challenge to Authorship Attribution (2021.eacl-main)

Copied to clipboard

Haining Wang, Allen Riddell, Patrick Juola

Challenge:	Existing studies on authorship attribution have shown that authorial style changes with respect to sentence length, word use, readability, and certain part-of-speech ratios.
Approach:	They propose to measure the effect of writing mode on authorial style in a corpus of documents composed online and offline using a traditional word processor.
Outcome:	The authors show that online writing differs from offline writing in terms of sentence length, word use, readability, and certain part-of-speech ratios.

Complex Word Identification as a Sequence Labelling Task (P19-1)

Copied to clipboard

Sian Gooding, Ekaterina Kochmar

Paraphrasing

Challenge:	Complex Word Identification (CWI) is a crucial first step in a simplification pipeline.
Approach:	They propose a system that performs CWI in context without extensive feature engineering and outperforms state-of-the-art systems on this task.
Outcome:	The proposed system outperforms state-of-the-art systems on complex word identification.

Combining Automatic Labelers and Expert Annotations for Accurate Radiology Report Labeling Using BERT (2020.emnlp-main)

Copied to clipboard

Akshay Smit, Saahil Jain, Pranav Rajpurkar, Anuj Pareek, Andrew Ng, Matthew Lungren

Fast and Accurate Reordering with ITG Transition RNN (C18-1)

Challenge:	Existing approaches to label radiology text reports rely on feature engineering based on medical domain knowledge or manual annotations by experts.
Approach:	They propose a BERT-based approach to medical image report labeling that exploits the scale of available rule-based systems and the quality of expert annotations.
Outcome:	The proposed model outperforms the previous best rules-based labeler with statistical significance on one of the largest datasets of chest x-rays.

Copied to clipboard

Hao Zhang, Axel Ng, Richard Sproat

Machine Translation

Challenge:	Attention-based sequence-to-sequence neural networks learn to jointly align and translate.
Approach:	They propose to use a reordering RNN that shares the input encoder with the decoder to decouple re-ordering from translation.
Outcome:	The proposed model can achieve superior reordering accuracy without feature engineering and is 2.5x faster in decoding.

A Review on Deep Learning Techniques Applied to Answer Selection (C18-1)

Copied to clipboard

Tuan Manh Lai, Trung Bui, Sheng Li

Question Answering

Challenge:	Existing deep learning methods for answer selection are not feature engineering or expensive external resources.
Approach:	They propose to use deep learning methods to analyze and predict answer quality . they use a set of candidate answers to identify which of the candidates answers the question correctly.
Outcome:	The proposed methods produce impressive performance without feature engineering or expensive external resources.

A Survey on Recent Advances in Named Entity Recognition from Deep Learning models (C18-1)

Copied to clipboard

Vikas Yadav, Steven Bethard

Information Extraction & Text Mining

Challenge:	Named Entity Recognition (NER) is a key component in NLP systems for question answering, information retrieval, relation extraction, etc.
Approach:	They propose to use recurrent neural networks to generate NERs over characters, sub-words and/or word embeddings to improve named entity recognition.
Outcome:	The proposed architectures are better than those based on feature engineering and other supervised or semi-supervised learning algorithms.

Affect inTweets: A Transfer Learning Approach (2020.lrec-1)

Copied to clipboard

Linrui Zhang, Hsin-Lun Huang, Yang Yu, Dan Moldovan

Sentence Level Human Translation Quality Estimation with Attention-based Neural Networks (2020.lrec-1)

Challenge:	Existing machine learning models require considerable effort to design task specific features to understand affectual states of people.
Approach:	They propose a transfer-learning based approach to infer the affectual state of a person from tweets.
Outcome:	The proposed model ranks 2nd, 4th and 6th in four of the four subtasks on SemEval-2018 task 1: Affect in Tweets.

Copied to clipboard

Yu Yuan, Serge Sharoff

Machine Translation

Challenge:	Existing methods for assessing translation quality rely on manual features and external knowledge.
Approach:	They propose to use a neural model without feature engineering to detect which parts in sentence pairs are most relevant for assessing quality.
Outcome:	The proposed model outperforms feature-based methods on a large human annotated dataset.

FeRG-LLM : Feature Engineering by Reason Generation Large Language Models (2025.findings-naacl)

Copied to clipboard

Jeonghyun Ko, Gyeongyun Park, Donghoon Lee, Kyunam Lee

Understanding Feature Focus in Multitask Settings for Lexico-semantic Relation Identification (2021.findings-acl)

Challenge:	FeRG-LLM is a large language model that performs feature engineering at an 8billion-parameter scale.
Approach:	They propose a framework to perform feature engineering at an 8billion-parameter scale using conversational dialogues.
Outcome:	The proposed framework outperforms Llama 3.1 70B and Llma 3.2 on most datasets while using fewer resources and achieving reduced inference time.

Copied to clipboard

Houssam Akhmouch, Gaël Dias, Jose G. Moreno

Responsible & Trustworthy NLP

Challenge:	Lexico-semantic relations embody symmetric and asymmetric linguistic phenomena such as synonymy (e.g. phone telephone), cohyponymy (, cohypoonymy, hypernymy, meronymy) and more can be enumerated.
Approach:	They propose to combine feature engineering and multitask architectures to identify lexico-semantic relations by combining asymmetric distributional features with shared-private models.
Outcome:	The proposed models improve over binary and fully-shared classifiers and balance the focus on features between private and shared layers 1 and 2 .

Multi-level Diagnosis and Evaluation for Robust Tabular Feature Engineering with Large Language Models (2025.findings-emnlp)

Copied to clipboard

Yebin Lim, Susik Yoon

Responsible & Trustworthy NLP

Challenge:	Recent advances in large language models have shown promise in feature engineering for tabular data, but reliability concerns persist due to variability in generated outputs.
Approach:	They propose a multi-level diagnosis and evaluation framework to assess the robustness of large language models in feature engineering across diverse domains.
Outcome:	The proposed framework assesses the robustness of large language models across domains.

Comparing Machine Learning and Deep Learning Approaches on NLP Tasks for the Italian Language (2020.lrec-1)

Copied to clipboard

Bernardo Magnini, Alberto Lavelli, Simone Magnolini

Information Extraction & Text Mining

Challenge:	Using available datasets, we compare deep learning and traditional machine learning methods for various NLP tasks in Italian.
Approach:	They compare deep learning and traditional machine learning methods for various NLP tasks in Italian.
Outcome:	The proposed methods outperform traditional methods in sequence tagging tasks and classification tasks in Italian.

Incorporating Word Attention into Character-Based Word Segmentation (N19-1)

Copied to clipboard

Shohei Higashiyama, Masao Utiyama, Eiichiro Sumita, Masao Ideuchi, Yoshiaki Oida, Yohei Sakamoto, Isaac Okada

Syntactic Text Processing

Challenge:	Word segmentation models are used to minimize the effort in feature engineering.
Approach:	They propose a character-based model that learns the importance of multiple candidate words for a corresponding character on the basis of an attention mechanism and makes use of it for segmentation decisions.
Outcome:	The proposed model outperforms the state-of-the-art models on Japanese and Chinese benchmark datasets.

Sanskrit Word Segmentation Using Character-level Recurrent and Convolutional Neural Networks (D18-1)

Copied to clipboard

Oliver Hellwig, Sebastian Nehrdich

Syntactic Text Processing

Challenge:	Using end-to-end neural network models, Sanskrit is tokenized by splitting compounds and resolving phonetic merges.
Approach:	They propose end-to-end neural network models that tokenize Sanskrit by jointly splitting compounds and resolving phonetic merges.
Outcome:	The proposed models outperform the state-of-the-art for the task of splitting compounds and resolving phonetic merges.

Can Models Help Us Create Better Models? Evaluating LLMs as Data Scientists (2026.findings-eacl)

Copied to clipboard

Michał Pietruszka, Łukasz Borchmann, Aleksander Jędrosz, Paweł Morawiecki

Challenge:	Current benchmarks assess LLMs on more isolated capabilities, such as language understanding and question-answering.
Approach:	They propose a benchmark to evaluate the ability of large language models (LLMs) to perform feature engineering.
Outcome:	The proposed benchmark evaluates the ability of large language models to perform feature engineering, a critical and knowledge-intensive task in data science.

Morphological disambiguation from stemming data (2020.coling-main)

Copied to clipboard

Antoine Nzeyimana

Syntactic Text Processing

Challenge:	morphologically rich languages require ambiguous analysis to be effective . morphology tools are limited for morphlogical analysis, disambiguation, and annotation .
Approach:	They propose to learn to morphologically disambiguate Kinyarwanda verbal forms from a crowd-sourced stemming dataset using feature engineering and a feed-forward neural network based classifier.
Outcome:	The proposed method achieves about 89% non-contextualized disambiguation accuracy from a crowd-sourced dataset.

CORT: A New Baseline for Comparative Opinion Classification by Dual Prompts (2022.findings-emnlp)

Copied to clipboard

Yequan Wang, Hengran Zhang, Aixin Sun, Xuying Meng

Low-Resource NLP

Challenge:	Comparative opinion classification is a common linguistic phenomenon.
Approach:	They propose a framework for comparative opinion classification using embedded knowledge in pre-trained language models.
Outcome:	The proposed framework delivers state-of-the-art and robust performance on all benchmark datasets.

Neural Automated Essay Scoring Incorporating Handcrafted Features (2020.coling-main)

Copied to clipboard

Masaki Uto, Yikuan Xie, Maomi Ueno

Representation Learning

Challenge:	Automated essay scoring (AES) relies on handcrafted features, but recent studies have proposed a hybrid method that integrates handcrafted essay-level features into a DNN-AES model.
Approach:	They propose a method that integrates handcrafted features into a DNN-AES model.
Outcome:	The proposed method significantly improves the accuracy of existing methods.

CoDet-M4: Detecting Machine-Generated Code in Multi-Lingual, Multi-Generator and Multi-Domain Settings (2025.findings-acl)

Copied to clipboard

Daniil Orel, Dilshod Azizov, Preslav Nakov

Programming Languages in NLP

Challenge:	Large Language Models (LLMs) have revolutionized code generation but have significant consequences for programming skills, ethics, and assessment integrity.
Approach:	They propose a framework capable of distinguishing between human-written and LLM-generated program code across multiple programming languages, code generators, and domains.
Outcome:	The proposed framework distinguishes between human-written and LLM-generated program code across multiple programming languages, code generators, and domains.

AssistedDS: Benchmarking How External Domain Knowledge Assists LLMs in Automated Data Science (2025.findings-emnlp)

Copied to clipboard

An Luo, Xun Xian, Jin Du, Fangqiao Tian, Ganghua Wang, Ming Zhong, Shengchun Zhao, Xuan Bi, Zirui Liu, Jiawei Zhou, Jayanth Srinivasa, Ashish Kundu, Charles Fleming, Mingyi Hong, Jie Ding

Robustness in NLP

Challenge:	Large language models (LLMs) have advanced the automation of data science workflows, yet it remains unclear whether they can critically leverage external domain knowledge as human data scientists do in practice.
Approach:	They propose a benchmark to evaluate how large language models handle external domain knowledge in tabular prediction tasks.
Outcome:	The proposed model evaluates whether it can critically leverage external domain knowledge as human data scientists do in practice.

SPIO: Ensemble and Selective Strategies via LLM-Based Multi-Agent Planning in Automated Data Science (2026.acl-long)

Copied to clipboard

Wonduk Seo, Juhyeon Lee, Yanjun Shao, Qingshan Zhou, Seunghyun Lee, Yi Bu

Understanding Structured Financial Data with LLMs: A Case Study on Fraud Detection (2026.acl-long)

Challenge:	Large Language Models (LLMs) have enabled dynamic reasoning in automated data analytics, but rigid, single-path workflows restrict strategic exploration and often lead to suboptimal outcomes.
Approach:	a new framework replaces rigid workflows with adaptive, multi-path planning . the framework offers two operating modes: SPIO-S and SPIO -E .
Outcome:	a new framework outperforms state-of-the-art pipelines on Kaggle and OpenML benchmarks.

Copied to clipboard

Xuwei Tan, Yao Ma, Xueru Zhang

Towards Interpretable Tabular Reasoning: Enhancing LLM Reasoning on Tabular Data with Pre-Constructed Logic Graph (2026.acl-long)

Challenge:	Large Language Models (LLMs) are expensive to develop and maintain and require extensive feature engineering to perform.
Approach:	They propose a two-stage approach that serializes a compact subset of numeric/categorical attributes into natural language and performs retrieval-augmented in-context learning over label-aware, instance-level exemplars.
Outcome:	The proposed approach significantly improves F1/MCC over direct prompting and is competitive with strong tabular baselines in several settings.

Copied to clipboard

Lirong Gao, Zewei Yu, Zhongrui Yin, Qi Zhang, Yuke Zhu, Bo Zheng, Haobo Wang, Junbo Zhao, Gang Chen, Sheng Guo