However, LoRA suffers from slow convergence. We introduce Dimension-Sharding Adaptation (DiSHA), which expands the PEFT design space to unlock lower intrinsic ranks and faster convergence by default. Within DiSHA's design space, we propose Block Affine Adaptation (Bone), a computationally efficient structure that delivers both high performance and efficiency. While certain DiSHA configurations may result in colinear updates to weight shards, we address this with Block Affine Transformation Adaptation (BAT), a nonlinear variant of DiSHA. BAT introduces nonlinearity by combining trainable matrices with original weight shards in a nonlinear manner, inducing nonlinearity in matrix updates without introducing additional parameters. Empirical results show that Bone, under the DiSHA framework, consistently outperforms LoRA variants in both NLG and NLU tasks, with significantly improved computational efficiency. Further analysis demonstrates that BAT enhances model capabilities by leveraging its nonlinear design. The emergence of Large Language Models (LLMs) has fundamentally transformed many traditional technologies Radford et al. (2019); Raffel et al. (2020).

Workplace sexual harassment continues to be one of the most under-reported crimes in India and the world over. A primary reason for this is the lack of awareness, both by'victims' as well as'perpetrators', on what constitutes sexual harassment, and the action to be taken when faced with it. Project Disha is here to help. This project consists of an Artificial Intelligence (AI) enabled chatbot at its core, which sits on top of an eLearning course on'Prevention of Sexual Harassment', and answers learner questions and alleviates concerns on the topic. This chatbot, called Disha the Learning Guide, also helps learners freely navigate the content without any restrictions and is the world's first chatbot powering a course developed in Articulate Storyline.