Think Big, Generate Quick: LLM-to-SLM for Fast Autoregressive Decoding
Paper • 2402.16844 • Published • 2
specimba
specimba
AI & ML interests
None yet
Recent Activity
updated a model about 6 hours ago
specimba/nexus-os-v2-dualmode published a model about 6 hours ago
specimba/nexus-os-v2-dualmode liked a model about 7 hours ago
mradermacher/qwen3.5-9b-nsfw-captioning-v5-GGUFOrganizations
upvoted a paper 1 day ago
upvoted a paper 4 days ago
upvoted an article 4 days ago
Article
Teaching Code Models to Write Secure Code: The SecureCode Collection
scthornton
• • 3 upvoted an article 6 days ago
Article
SmolLM - blazingly fast and remarkably powerful
- +1
loubnabnl, anton-l, eliebak
• • 455 upvoted 5
papers 6 days ago
Paper Espresso: From Paper Overload to Research Insight
Paper • 2604.04562 • Published • 13
TriSpec: Ternary Speculative Decoding via Lightweight Proxy Verification
Paper • 2601.23180 • Published • 1
MARS: Unleashing the Power of Speculative Decoding via Margin-Aware Verification
Paper • 2601.15498 • Published • 1
P-EAGLE: Parallel-Drafting EAGLE with Scalable Training
Paper • 2602.01469 • Published • 4
PACER: Blockwise Pre-verification for Speculative Decoding with Adaptive Length
Paper • 2602.01274 • Published • 1
upvoted 2
papers 7 days ago
upvoted a collection 7 days ago
upvoted 6
papers 7 days ago
Greedy Multi-Path Block Verification for Faster Decoding in Speculative Sampling
Paper • 2602.16961 • Published • 1
Dynamic Delayed Tree Expansion For Improved Multi-Path Speculative Decoding
Paper • 2602.16994 • Published • 1
TAPS: Task Aware Proposal Distributions for Speculative Sampling
Paper • 2603.27027 • Published • 144
Accelerating Speculative Decoding with Block Diffusion Draft Trees
Paper • 2604.12989 • Published • 8
DFlash: Block Diffusion for Flash Speculative Decoding
Paper • 2602.06036 • Published • 80
S2D2: Fast Decoding for Diffusion LLMs via Training-Free Self-Speculation
Paper • 2603.25702 • Published • 8
upvoted 2
papers 10 days ago