Publications/Natural Language Processing & LLMs for Low-Resource Languages
Workshop · 2026
Kahaani: A Multimodal Co-Creative Storytelling System
Samee Arif, Muhammad Saad Haroon, Aamina Jamal Khan, Taimoor Arif, Agha Ali Raza, Awais Athar
EACL 2026 Student Research Workshop, Rabat, Morocco
Summary
Kahaani is a storytelling system in which a child and an AI build a story together. It combines a language model with generated speech, music, and video, and structures the collaboration around two classical storytelling frameworks, Freytag's Pyramid and Propp's narrative functions. The goals are to help children improve their English, learn life lessons through story morals, and understand how stories are built, all in a fun and engaging way. We evaluate each AI component and report a user study with three parent-child pairs on the overall experience and educational value.
Cite this paper
Samee Arif, Muhammad Saad Haroon, Aamina Jamal Khan, Taimoor Arif, Agha Ali Raza, Awais Athar. Kahaani: A Multimodal Co-Creative Storytelling System. EACL 2026 Student Research Workshop, Rabat, Morocco.
@inproceedings{arif2026kahaani,
author = {Samee Arif and Muhammad Saad Haroon and Aamina Jamal Khan and Taimoor Arif and Agha Ali Raza and Awais Athar},
title = {{Kahaani: A Multimodal Co-Creative Storytelling System}},
booktitle = {EACL 2026 Student Research Workshop, Rabat, Morocco},
year = {2026},
url = {https://aclanthology.org/2026.eacl-srw.25/}
}
- IRTIQA: Evolving in English, Evaluated in Urdu — Does Agent Self-Evolution Transfer Across Languages?Findings of EACL 2027, Athens, Greece
- UrduCultureQA: A Culturally Grounded Benchmark for Urdu Question Answering and Alignment EvaluationFindings of EACL 2027, Athens, Greece
- Pak3H: Evaluating the Cost of Cultural Mismatch in LLM Alignment with a Human-Contextualized Urdu BenchmarkEMNLP 2026 (Main), Budapest, Hungary
- Evaluating Robustness of LLM-Based Ambient Scribes for SOAP Note GenerationCHIL 2026 (Conference on Health, Inference, and Learning), PMLR vol. 333
- Large Language Models Show Dunning-Kruger-Like Effects in Multilingual Fact-CheckingScientific Reports 16, 7594 (2026), a Nature Portfolio journal