pith. sign in

Yongji Wu

Identifiers

  • name variant Yongji Wu 0.60 · backfill

Papers (4)

  1. AstraFlow: Dataflow-Oriented Reinforcement Learning for Agentic LLMs cs.LG · 2026 · author #6
  2. STS: Efficient Sparse Attention with Speculative Token Sparsity cs.LG · 2026 · author #3
  3. Foundry: Template-Based CUDA Graph Context Materialization for Fast LLM Serving Cold Start cs.DC · 2026 · author #2
  4. RLBoost: Harvesting Preemptible Resources for Cost-Efficient Reinforcement Learning on LLMs cs.DC · 2025 · author #1

Mentions

  • 2605.15508 #3 · arxiv_oai · confidence 0.70 Yongji Wu
  • 2605.15565 #6 · arxiv_oai · confidence 0.70 Yongji Wu

Frequent Coauthors