Zhi Wei Tan

zwtan-cs's activity

• All
• Models
• Datasets
• Spaces
• Buckets
• Papers
• Collections
• Community
• Posts
• Upvotes
• Likes
• Articles

upvoted 2 papers about 18 hours ago

DLite: History-free Speculative Decoding for Agentic AI

Paper • 2610.23787 • Published 11 days ago • 36

Meta Expert: Cross-Layer Expert Sharing for Storage- and Retrieval-Efficient MoE

Paper • 2610.27598 • Published 7 days ago • 45

liked 3 Spaces 3 days ago

Running

4

Awesome KV Cache Optimization

🚀

4

System-Aware KV Cache Optimization for LLM Serving [Survey]

Running

1

Repro - Speculative Coupled Decoding for Training-Free Lossless Acceleration of Autoregressive Visual Generation

🎯

1

Collaborate on research logs with a coding agent

Running

Agents

4

Speculative Decoding Demo

🚀

4

Generate text with speculative decoding

upvoted a paper 3 days ago

Cache Engineering for Agents

Paper • 2610.28577 • Published 7 days ago • 159

liked 2 Spaces 4 days ago

Paused

MCP

3

Context KV Cache

🧩

3

Read KV cache entries from task context

Running

2

KV Cache Budget Simulator

📐

2

KV cache bytes, OOM walls, PagedAttention, Pareto frontier

upvoted 2 papers 4 days ago

Large Flow Language Models

Paper • 2610.28559 • Published 7 days ago • 24

Specter: Fast MoE Inference via Quantization-Enabled Multi-Step Speculative Offloading

Paper • 2511.14102 • Published 7 days ago • 27

upvoted 3 papers 5 days ago

Factorization-Guided Training and Inference for Diffusion-Drafted Speculative Decoding

Paper • 2610.23935 • Published 20 days ago • 16

MemT-1.0: Towards Memory-Native Evolution

Paper • 2610.26825 • Published 8 days ago • 181

Self-Evaluating Recursive Agents

Paper • 2610.04902 • Published 23 days ago • 17

upvoted 2 papers 6 days ago

Virtual Head Attention

Paper • 2610.17959 • Published 14 days ago • 25

SCOPE: Semantic Bounded Credit Optimization for Prompt Engineering in Text-to-image Generation

Paper • 2610.22264 • Published 12 days ago • 51

liked a dataset 7 days ago

vimalnakrani/stillwarm-kv-cache-artifact

Viewer • Updated Jul 6 • 1 • 55 • 8

upvoted 3 papers 7 days ago

KV4-RL: Unlocking Faster Asynchronous Rollouts for RLVR with INT4 KV Cache Quantization

Paper • 2610.14448 • Published 17 days ago • 49

History Replays Itself: Unleashing Long-Context Capability in Linear Attention

Paper • 2610.15715 • Published 16 days ago • 26

Towards Agentic Robots: Model, Data, Environment, and Harness

Paper • 2610.20442 • Published 13 days ago • 43

liked a dataset 10 days ago

emgena/speculative_decoding_draft_guard_teaser

Viewer • Updated 13 days ago • 50 • 98 • 8