Papers with Education
MICo: Preventative Detoxification of Large Language Models through Inhibition Control (2024.findings-naacl)
Copied to clipboard
Roy Siegelmann, Ninareh Mehrabi, Palash Goyal, Prasoon Goyal, Lisa Bauer, Jwala Dhamala, Aram Galstyan, Rahul Gupta, Reza Ghanadan
| Challenge: | Large Language Models (LLMs) have a tendency to devolve into toxic degeneration . model may classify prompts as toxic or non-toxic and categorically refuse to respond to those deemed toxic. |
| Approach: | They propose a mechanism for LLM detoxification by labeling acceptable and unacceptable examples and including a corresponding acceptable rewrite with every unacceptable example. |
| Outcome: | The proposed model improves on the baseline model and shows that it detects and rewrites toxic and harmful examples. |
A Reinforcement Learning Approach to Improve Low-Resource Machine Translation Leveraging Domain Monolingual Data (2024.lrec-main)
Copied to clipboard
| Challenge: | Existing methods for fine-tuning domain adaptation have overfitting problem in low-resource domains . lack of parallel data makes it difficult for model to learn domain-specific knowledge . |
| Approach: | They propose a Reinforcement Learning Domain Adaptation method for Neural Machine Translation that uses in-domain source monolingual data to make up for the lack of parallel data. |
| Outcome: | The proposed method can alleviate overfitting and reinforce the model to learn domain-specific knowledge. |
GAIfE: Using GenAI to Improve Literacy in Low-resourced Settings (2025.findings-naacl)
Copied to clipboard
Allahsera Auguste Tapo, Nouhoum Coulibaly, Seydou Diallo, Sebastien Diarra, Christopher M Homan, Mamadou K. Keita, Michael Leventhal
| Challenge: | Illiteracy is a predictor of many negative social and personal outcomes in underresourced countries, where few books exist that are suitable for children to learn to read from. |
| Approach: | They propose to use generative AI to create culturally-engaging materials for learning in mali's vehicular language Bambara by multiplying the content by 10 times . authors propose to apply bias-aware tools to reduce illiteracy and improve learning outcomes through native language education. |
| Outcome: | The proposed toolchain and workflow can be adapted to address low literacy in mali using generative AI. |
Incorporating Word-level Phonemic Decoding into Readability Assessment (2024.lrec-main)
Copied to clipboard
| Challenge: | a recent study suggests that automatic readability assessment is not able to provide interpretability for teachers and educators. |
| Approach: | They propose to incorporate phonetic and orthographic features into automatic readability assessment by handcrafted feature sets. |
| Outcome: | a new feature set shows comparable performance to larger feature sets on grade-level classification in english . authors say the model improves on multiple readability datasets but lacks interpretability . |
EduVidQA: Generating and Evaluating Long-form Answers to Student Questions based on Lecture Videos (2025.emnlp-main)
Copied to clipboard
| Challenge: | This paper explores using Multimodal Large Language Models (MLLMs) to respond to student questions from online lectures . MLLM is a novel question answering task of real world significance . |
| Approach: | They propose to use Multimodal Large Language Models to automatically respond to student questions from online lectures by using a dataset of 5252 question-answer pairs from 296 computer science videos. |
| Outcome: | The proposed model can fine tune and fine tune questions from 296 computer science videos and show that students' preferences are important to the task. |