Trained early exit head to be used with Dynamic Self-Speculative Decoding
🤝 Open to Collab
Florian Valade
FlorianVal
AI & ML interests
Efficiency in Deep learning
Recent Activity
liked a model 20 days ago
zai-org/GLM-5.3 upvoted an article 2 months ago
Tensor Parallelism (TP) in Transformers: 5 Minutes to Understand updated a Space 6 months ago
FlorianVal/LaMaisonOrganizations
