What we’re about
You are in love with Data, Artificial Intelligence, Machine learning, Big Data or IoT so join us to learn about the cutting edge of AI! \
Whether you are in computer science, mathematics, statistics, management, marketing, etc. If you think this activity is for scientists, change your mind by joining us. We will prove to you that AI is easy. Welcome to the new world.
Upcoming events (3)
See all- [RG] Interactive Debugging and Steering of Multi-Agent AI SystemsLink visible for attendees
Our host Dr Abdoullahi Diasse, Senior AI Engineer at Syrate Sarl, will present the paper Vision as LoRA whose abstract follows:
We introduce Vision as LoRA (VoRA), a novel paradigm for transforming an LLM into an MLLM. Unlike prevalent MLLM architectures that rely on external vision modules for vision encoding, VoRA internalizes visual capabilities by integrating vision-specific LoRA layers directly into the LLM. This design allows the added parameters to be seamlessly merged into the LLM during inference, eliminating structural complexity and minimizing computational overhead. Moreover, inheriting the LLM's ability of handling flexible context, VoRA can process inputs at arbitrary resolutions.
To further strengthen VoRA's visual capabilities, we introduce a block-wise distillation method that transfers visual priors from a pre-trained ViT into the LoRA layers, effectively accelerating training by injecting visual knowledge. Additionally, we apply bi-directional attention masks to better capture the context information of an image. We successfully demonstrate that with additional pre-training data, VoRA can perform comparably with conventional encode-based MLLMs.
![[RG] Interactive Debugging and Steering of Multi-Agent AI Systems cover photo](https://secure.meetupstatic.com/photos/event/b/a/e/3/event_528047843.webp "[RG] Interactive Debugging and Steering of Multi-Agent AI Systems")
