Shakti-VLMs: Scalable Vision-Language Models for Enterprise AI
This provides an efficient solution for enterprise-scale multimodal tasks by reducing data requirements through architectural and training innovations.
The authors tackled data efficiency challenges in multimodal learning by introducing Shakti VLM, a family of vision-language models with 1B and 4B parameters, achieving competitive results in document understanding, visual reasoning, OCR extraction, and general multimodal reasoning with fewer tokens than existing methods.
We introduce Shakti VLM, a family of vision-language models in the capacity of 1B and 4B parameters designed to address data efficiency challenges in multimodal learning. While recent VLMs achieve strong performance through extensive training data, Shakti models leverage architectural innovations to attain competitive results with fewer tokens. Key advancements include QK-Normalization for attention stability, hybrid normalization techniques, and enhanced positional encoding. A three-stage training strategy further optimizes learning efficiency. Evaluations show that Shakti-Shakti-VLM-1B and Shakti-VLM-4B excel in document understanding, Visual Reasoning, OCR extraction, and general multimodal reasoning. Our results highlight that high performance can be achieved through model design and training strategy rather than sheer data volume, making Shakti an efficient solution for enterprise-scale multimodal tasks.