Cross-paradigm speculative decoding method to accelerate LLM inference (Kwon et al., Findings of EMNLP 2026).
AI & ML interests
Cross-Paradigm Speculative Decoding
Recent Activity
View all activity
models 8
WhiFlash/Qwen3-4B-NeuralRouter-ddt-e3-b128
Updated • 42
WhiFlash/Qwen3-4B-NeuralRouter-ddt-e3-b32
Updated • 44
WhiFlash/Qwen3-4B-NeuralRouter-df-e3
Updated • 47
WhiFlash/Qwen3-8B-NeuralRouter-ddt-e3-b128
Updated • 43
WhiFlash/Qwen3-8B-NeuralRouter-ddt-e3-b32
Updated • 50
WhiFlash/Qwen3-8B-NeuralRouter-df-e3
Updated • 50
WhiFlash/Qwen3-8B-EAGLE3
0.4B • Updated • 159
WhiFlash/Qwen3-4B-EAGLE3
0.2B • Updated • 190
datasets 0
None public yet