pith. the verified trust layer for science. sign in

Chi-Chih Chang

Identifiers

No identifiers captured yet.

Papers (3)

  1. DisagMoE: Computation-Communication overlapped MoE Training via Disaggregated AF-Pipe Parallelism cs.LG · 2026 · author #2
  2. DARE: Diffusion Language Model Activation Reuse for Efficient Inference cs.LG · 2026 · author #4
  3. Faster LLM Inference via Sequential Monte Carlo cs.LG · 2026 · author #3

Mentions

No mention provenance yet.

Frequent Coauthors