Hugging Face's logo
Bojan Kostic's picture
3

Bojan Kostic

bojan2501
·

AI & ML interests

None yet

Recent Activity

reacted to chansung's post with 👍 9 days 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.
commented on an article 9 days ago
Open-R1: a fully open reproduction of DeepSeek-R1
liked a model 20 days ago
danielheinz/e5-base-sts-en-de
View all activity

Organizations

None yet

bojan2501's activity

liked a model 20 days ago
liked a model 7 months ago
liked a model 11 months ago