- Rating
- 3
- Confidence
- 2
- Informativeness
- 2
- Presentation
- 2
- Soundness
- 3
Read review
Paper summary
This paper studies multiversion MoE serving, where customized versions of a shared backbone must be stored compactly and executed jointly. The authors propose REPS, a compressed expert representation. It encodes weights without fitting a codebook and reconstructs them without codeword lookup. Experiments show that REPS matches AQLM-level quality while reducing TTFT and weight allocation.
Critical strengths
-
The paper addresses multiversion MoE serving under GPU memory constraints, an important problem, and solves it effectively.
-
The evaluation is rigorous
Critical weaknesses
- At equal budgets, AQLM achieves lower PPL; REPS's quality advantage comes from a larger continuation budget. AQLM also has lower ITL than REPS. REPS therefore yields neither a quality nor an overall throughput advantage — its gains are limited to encoding speed and prefill TTFT
Questions for authors
- Could you provide an experimental or analytical comparison against ExpertWeave at the same number of versions and the same workloads?