# Cache-Aware Joint Router Adaptation for Memory-Efficient MoE Inference

arXiv: https://arxiv.org/abs/2609.04895 · HTML: https://mlsanity.com/paper/2609.04895/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Placed in Expert routing strategies, one of 12 directions in Mixture-of-experts.
- Builds on 14 of its 21 references that are in the map.
- First posted on arXiv 4 Sep 2026.

## Placed in

- Mixture-of-experts / [Expert routing strategies](https://mlsanity.com/moe/expert-routing-strategies/)

## Builds on (most cited first, 14 of 14)

- [Training Verifiers to Solve Math Word Problems](https://mlsanity.com/paper/2110.14168/) (2021)
- [Efficient Memory Management for Large Language Model Serving with PagedAttention](https://mlsanity.com/paper/2309.06180/) (2023)
- [Measuring Mathematical Problem Solving With the MATH Dataset](https://mlsanity.com/paper/2103.03874/) (2021)
- [CommonsenseQA: A Question Answering Challenge Targeting Commonsense Knowledge](https://mlsanity.com/paper/1811.00937/) (2018)
- [gpt-oss-120b & gpt-oss-20b Model Card](https://mlsanity.com/paper/2508.10925/) (2025)
- [FlexGen: High-Throughput Generative Inference of Large Language Models with a Single GPU](https://mlsanity.com/paper/2303.06865/) (2023)
- [DeepSpeed-MoE: Advancing Mixture-of-Experts Inference and Training to Power Next-Generation AI Scale](https://mlsanity.com/paper/2201.05596/) (2022)
- [Tutel: Adaptive Mixture-of-Experts at Scale](https://mlsanity.com/paper/2206.03382/) (2022)
- [ProMoE: Fast MoE-based LLM Serving using Proactive Caching](https://mlsanity.com/paper/2410.22134/) (2024)
- [Taming Latency-Memory Trade-Off in MoE-Based LLM Serving via Fine-Grained Expert Offloading](https://mlsanity.com/paper/2502.05370/) (2025)
- [SpecMD: A Comprehensive Study On Speculative Expert Prefetching](https://mlsanity.com/paper/2602.03921/) (2026)
- [Temporally Extended Mixture-of-Experts Models](https://mlsanity.com/paper/2604.20156/) (2026)
- [A Spatio-Temporal Expert Prefetching Framework for Efficient MoE-based LLM Inference](https://mlsanity.com/paper/2606.15453/) (2026)
- [MoEless: Efficient MoE LLM Serving via Serverless Computing](https://mlsanity.com/paper/2603.06350/) (2026)

## Built on by (most cited first, 3 of 3)

- [Efficient Mixture-of-Experts with Speculative Decoding via Expert Coactivation](https://mlsanity.com/paper/2609.22471/) (2026)
- [MaskCoFT: Masked Co-Adaptive Fine-Tuning for Memory-Efficient MoE Inference](https://mlsanity.com/paper/2609.34077/) (2026)
- [PA1-OPD: One-Step Prefix-Augmented On-Policy Distillation for Language Model Pretraining](https://mlsanity.com/paper/2610.19601/) (2026)

## Abstract (arXiv)

Mixture-of-Experts (MoE) models activate few experts per token, yet their full expert sets can exceed GPU memory and require repeated weight transfers during decoding. We formulate expert-cache management as a model-side algorithmic problem and propose cache-aware post-training that jointly adapts the MoE backbone and lightweight auxiliary routers while preserving the native inference-time Top-K rule. The update-only Temporal Router learns same-layer retention across tokens without proactive loading. The full Spatio-Temporal Router adds a Spatio Router that uses the causal predecessor's hidden state to refine the temporal cache before target-layer access. We evaluate both modes on Qwen3 and GPT-OSS across GSM8K, MATH, and CommonsenseQA. Temporal Router consistently improves hit rate and reduces expert-weight traffic over matched LM-only baselines. On Qwen3, the full mode improves adjusted hit rate by 1.15--18.03 points and reduces traffic by 4.6--53.3% relative to the strongest evaluated prefetching baseline; GPT-OSS results are competitive but task-dependent. Auxiliary-only training preserves baseline accuracy but yields modest coverage gains; joint post-training achieves substantially higher coverage. Sensitivity analyses distinguish the effects of cache capacity, refinement budget, and cache-loss weight on coverage, traffic, and quality.