Han-Bit Kang's picture

110 25

Han-Bit Kang

hbkang

·

AI & ML interests

ML

Recent Activity

updated a collection 10 days ago

upvoted a paper 10 days ago

Softpick: No Attention Sink, No Massive Activations with Rectified Softmax

updated a collection 10 days ago

interesting architecture

View all activity

Organizations

None yet

hbkang's activity

updated a collection 10 days ago

cool-papers

40 items • Updated 10 days ago

upvoted a paper 10 days ago

Softpick: No Attention Sink, No Massive Activations with Rectified Softmax

Paper • 2504.20966 • Published 13 days ago • 26

updated 2 collections 10 days ago

interesting architecture

15 items • Updated 10 days ago • 1

talking-head-generation

24 items • Updated 10 days ago

upvoted a paper 10 days ago

Generative AI for Character Animation: A Comprehensive Survey of Techniques, Applications, and Future Directions

Paper • 2504.19056 • Published 15 days ago • 15

updated a collection 13 days ago

cool-papers

40 items • Updated 10 days ago

upvoted a paper 13 days ago

Group Downsampling with Equivariant Anti-aliasing

Paper • 2504.17258 • Published 18 days ago • 8

updated a collection 14 days ago

generative-model-training

12 items • Updated 14 days ago

upvoted a paper 14 days ago

Boosting Generative Image Modeling via Joint Image-Feature Synthesis

Paper • 2504.16064 • Published 20 days ago • 14

updated a collection 17 days ago

ID-Preserving Generation

5 items • Updated 17 days ago

upvoted a paper 17 days ago

DreamID: High-Fidelity and Fast diffusion-based Face Swapping via Triplet ID Group Learning

Paper • 2504.14509 • Published 22 days ago • 50

updated a collection 21 days ago

cool-papers

40 items • Updated 10 days ago

upvoted a paper 21 days ago

Does Reinforcement Learning Really Incentivize Reasoning Capacity in LLMs Beyond the Base Model?

Paper • 2504.13837 • Published 24 days ago • 122

upvoted a paper 25 days ago

Cobra: Efficient Line Art COlorization with BRoAder References

Paper • 2504.12240 • Published 26 days ago • 28

upvoted a paper 27 days ago

Seaweed-7B: Cost-Effective Training of Video Generation Foundation Model

Paper • 2504.08685 • Published about 1 month ago • 123

updated a collection about 1 month ago

cool-papers

40 items • Updated 10 days ago

upvoted a paper about 1 month ago

DDT: Decoupled Diffusion Transformer

Paper • 2504.05741 • Published Apr 8 • 73

updated a collection about 1 month ago

interesting architecture

15 items • Updated 10 days ago • 1

upvoted 2 papers about 1 month ago

TransMamba: Flexibly Switching between Transformer and Mamba

Paper • 2503.24067 • Published Mar 31 • 20

SmolVLM: Redefining small and efficient multimodal models

Paper • 2504.05299 • Published Apr 7 • 181