Papers by Tamer Elsayed
MAPLE: A Meta-learning Framework for Cross-Prompt Essay Scoring (2026.findings-acl)
Copied to clipboard
| Challenge: | Current approaches to automate essay scoring (AES) treat each writing task as a separate task, resulting in inconsistent performance. |
| Approach: | They propose a meta-learning framework that leverages prototypical networks to learn transferable representations across different writing prompts. |
| Outcome: | The proposed framework outperforms baseline models on ELLIPSE and ASAP (English) and LAILA (Arabic) on three diverse datasets. |
IDRISI-RA: The First Arabic Location Mention Recognition Dataset of Disaster Tweets (2023.acl-long)
Copied to clipboard
| Challenge: | a low resource language such as Arabic is understudied for geolocation extraction . a recent study found that geolocation is underutilized for low resource languages such as arabic . |
| Approach: | They propose a publicly-available Arabic Location Mention Recognition dataset . it provides human- and automatically-labeled versions of tweets in order of thousands and millions of tweet . |
| Outcome: | The proposed dataset provides human- and automatically-labeled versions in order of thousands and millions of tweets. |
LAILA: A Large Trait-Based Dataset for Arabic Automated Essay Scoring (2026.eacl-long)
Copied to clipboard
May Bashendy, Walid Massoud, Sohaila Eltanbouly, Salam Albatarni, Marwan Sayed, Abrar Abir, Houda Bouamor, Tamer Elsayed
| Challenge: | Existing Arabic resources are small in scale and lack trait-specific annotations. |
| Approach: | They propose to use LAILA to build a large Arabic AES dataset with holistic and trait-specific annotations of seven writing proficiency traits. |
| Outcome: | The LAILA dataset comprises 7,859 essays annotated with holistic and trait-specific scores on seven dimensions: relevance, organization, vocabulary, style, development, mechanics, and grammar. |
DART: A Large Dataset of Dialectal Arabic Tweets (L18-1)
Copied to clipboard
| Challenge: | The Arabic language is the fifth most widely spoken language in the world; more than 380 million people speak and write in Arabic. |
| Approach: | They propose to build a large manually-annotated multi-dialect dataset of Arabic tweets that is publicly available. |
| Outcome: | The proposed dataset is well-balanced over five main Arabic dialects: Egyptian, Maghrebi, Levantine, Gulf, and Iraqi. |
Can Large Language Models Automatically Score Proficiency of Written Essays? (2024.lrec-main)
Copied to clipboard
| Challenge: | Automated essay scoring (AES) is one of the earliest research problems in natural language processing. |
| Approach: | They propose to use large language models to analyze and score written essays using four different prompts. |
| Outcome: | The proposed models show comparable performance on four different prompts and a slight advantage over the state-of-the-art models. |
TRATES: Trait-Specific Rubric-Assisted Cross-Prompt Essay Scoring (2025.findings-acl)
Copied to clipboard
| Challenge: | Automated Essay Scoring (AES) has seen significant progress in assessing writing ability and trait scoring. |
| Approach: | They propose a trait-specific and rubric-based cross-prompt AES framework that is generic yet specific to the underlying trait. |
| Outcome: | The proposed framework achieves state-of-the-art across all traits on a widely-used dataset, with the generated LLM-based features being the most significant. |
Are We Ready for this Disaster? Towards Location Mention Recognition from Crisis Tweets (2020.coling-main)
Copied to clipboard
| Challenge: | Despite the widespread use of Twitter during emergencies, the majority of tweets do not have geoinformation. |
| Approach: | They propose to use Twitter to train location mention recognition models using different training settings. |
| Outcome: | The results show that training on near or far-away events boosts the performance compared to training on distant events. |
Qayyem: A Real-time Platform for Scoring Proficiency of Arabic Essays (2026.acl-demo)
Copied to clipboard
| Challenge: | Existing Arabic writing technologies primarily use a single quality score for essays, but there is limited support for Arabic AES. |
| Approach: | They propose a Web-based platform that integrates Arabic AES workflows with a user-friendly interface. |
| Outcome: | The proposed system integrates with existing Arabic scoring systems and provides a user-friendly interface. |