Bojan Kostic's picture

1 1 3

Bojan Kostic

bojan2501

·

AI & ML interests

None yet

Recent Activity

new activity 2 days ago

Qwen/Qwen3-30B-A3B:MoE question

upvoted a collection about 2 months ago

reacted to chansung's post with 👍 3 months ago

Simple summary on DeepSeek AI's Janus-Pro: A fresh take on multimodal AI! It builds on its predecessor, Janus, by tweaking the training methodology rather than the model architecture. The result? Improved performance in understanding and generating multimodal data. Janus-Pro uses a three-stage training strategy, similar to Janus, but with key modifications: ✦ Stage 1 & 2: Focus on separate training for specific objectives, rather than mixing data. ✦ Stage 3: Fine-tuning with a careful balance of multimodal data. Benchmarks show Janus-Pro holds its own against specialized models like TokenFlow XL and MetaMorph, and other multimodal models like SD3 Medium and DALL-E 3. The main limitation? Low image resolution (384x384). However, this seems like a strategic choice to focus on establishing a solid "recipe" for multimodal models. Future work will likely leverage this recipe and increased computing power to achieve higher resolutions.

View all activity

Organizations

None yet

bojan2501's activity

upvoted a collection about 2 months ago

QwQ

Qwen with Questions • 6 items • Updated 3 days ago • 94