[BEA @ ACL 2023] General-purpose tool for linguistic features extraction; Tested on readability assessment, essay scoring, fake news detection, hate speech detection, etc.

A corpus of short answers written by learners of English and graded with CEFR levels

The repository for the official website of the ACL Special Interest Group on Building Educational Applications. Contributions are welcome!

Official repo for "Improving Reading Comprehension Question Generation with Data Augmentation and Overgenerate-and-rank" (Outstanding paper at BEA@ACL 2023)

[ACL23, BEA workshop] Evaluating Reading Comprehension Exercises Generated by LLMs: A Showcase of ChatGPT in Education Applications

A Dataset of Japanese Lexical Complexity for Non-Native Readers

Code for our BEA 2024 Submission: Using Adaptive Empathetic Responses for Teaching English

BEA @ NAACL24 submission on grammar-controlled educational text generation

A CEFR-Graded Lexical Resource for Dutch as a Foreign Language Linked to Open Dutch WordNet

The repository contains detokenized GEC datasets.

Code for reproducing the GEC models from the "Adapting LLMs for Minimal-edit Grammatical Error Correction" paper.

Code for “Noise Steering for Controlled Text Generation,” exploring inference-time noise injection to increase diversity while maintaining quality and constraint adherence. (BEA @ ACL 26')

Code repository associated with "Noah-Manuel Michael and Andrea Horbach (2025). GermDetect: Verb Placement Error Detection Datasets for Learners of Germanic Languages. In Proceedings of the 20th Workshop on Building Educational Applications (BEA 2025). Association for Computational Linguistics."

Dataset and analysis code for BEA2025 paper @ ACL: "Alignment Drift in CEFR-prompted LLMs for Interactive Spanish Tutoring" (Almasi & Kristensen-McLachlan, 2025)

Repository for: Automated L2 Proficiency Scoring: Weak Supervision, Large Language Models, and Statistical Guarantees

Generating Better Items for Cognitive Assessments Using Large Language Models

Tools for working with UD treebanks of learner texts (and parallel treebanks in general).