Mahfuz Ahmed Anik, Abdur Rahman, Azmine Toushik Wasi, Md Manjurul Ahsan. Proceedings of the 1st Workshop on Language Models for Underserved Communities (LM4UC 2025). 2025.
Sense of ownership in writing confines our investment of thoughts, time, and contribution, leading to attachment to the output. However, using writing assistants introduces a mental dilemma, as some content isn’t directly our creation. For instance, we tend to credit Large Language Models (LLMs) more in creative tasks, even though all tasks are equal for them. Additionally, while we may not claim complete ownership of LLM-generated content, we freely claim authorship. We conduct a short survey to examine these issues and understand underlying cognitive processes in order to gain a better knowledge of human-centered aspects in writing and improve writing aid systems.
Preserving linguistic dialects is paramount for preserving cultural identity, fostering diversity, and maintaining cultural prestige. In the context of Bengali, the language exhibits religious distinctions through its two major dialects: the West Bengal (Hindu majority) and Bangladesh (Muslim majority) dialects, deeply rooted in the historical evolution of Bangla. Understanding and respecting these dialect-based nuances is crucial for Human-AI collaborative tools to generate culturally appropriate content and effectively assist human writers without limiting their creativity. To address these challenges, we introduce DiaFrame, a comprehensive framework tailored for understanding and tuning dialects in human-AI collaborative creative writing spaces, focusing on Bengali. It integrates advanced components such as active learning, real-time processing of human feedback, memory management, and contextual understanding to enhance user experience and ensure culturally sensitive content generation. Our contributions include providing a holistic end-to-end solution that enables the AI model to actively learn, adapt, and generate content aligned with user preferences and dialectical variations, ultimately fostering effective collaboration between humans and AI in creative writing endeavors.